Yelp_NLP_ABSA_Maureen.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/EAName/yelp-reviews/blob/main/Yelp_NLP_ABSA_Maureen.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "WsePP5VZuxQq"
      },
      "source": [
        "## Imports "
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "metadata": {
        "id": "92Dg33FmuxQs"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import os\n",
        "import random\n",
        "import numpy as np\n",
        "import pickle \n",
        "import re,string\n",
        "\n",
        "import nltk\n",
        "from nltk.corpus import stopwords\n",
        "from nltk.stem import PorterStemmer\n",
        "from nltk.stem import WordNetLemmatizer\n",
        "from nltk.tokenize import word_tokenize \n",
        "\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "from sklearn.cluster import KMeans, AgglomerativeClustering\n",
        "from sklearn.metrics.pairwise import cosine_similarity\n",
        "from sklearn.decomposition import TruncatedSVD, LatentDirichletAllocation\n",
        "from sklearn.svm import SVC\n",
        "from sklearn.preprocessing import LabelEncoder, OneHotEncoder\n",
        "from sklearn.ensemble import RandomForestClassifier\n",
        "from sklearn.metrics import roc_auc_score, accuracy_score, confusion_matrix, silhouette_score\n",
        "from sklearn.metrics import precision_score, recall_score, f1_score, classification_report\n",
        "from sklearn.metrics import mean_squared_error as MSE\n",
        "from sklearn.metrics import mean_absolute_error\n",
        "from sklearn.model_selection import cross_val_score, StratifiedKFold\n",
        "from sklearn.tree import DecisionTreeClassifier\n",
        "from sklearn.naive_bayes import MultinomialNB\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.ensemble import RandomForestRegressor\n",
        "from sklearn.metrics import accuracy_score\n",
        "from sklearn.model_selection import train_test_split, KFold\n",
        "from sklearn.manifold import MDS\n",
        "from sklearn.metrics.pairwise import cosine_similarity\n",
        "from sklearn.ensemble import GradientBoostingClassifier\n",
        "from sklearn.svm import LinearSVC\n",
        "\n",
        "\n",
        "\n",
        "from sklearn.model_selection import train_test_split \n",
        "from sklearn.model_selection import GridSearchCV \n",
        "from sklearn.model_selection import cross_val_score \n",
        "from sklearn.model_selection import StratifiedKFold\n",
        "from sklearn.model_selection import KFold \n",
        "from sklearn.model_selection import cross_val_predict\n",
        "\n",
        "\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "from gensim.models import Word2Vec,LdaMulticore, TfidfModel, CoherenceModel\n",
        "from gensim import corpora\n",
        "from gensim import similarities\n",
        "from gensim.models.doc2vec import Doc2Vec, TaggedDocument\n",
        "from gensim.models import LsiModel,LdaModel\n",
        "\n",
        "from keras.preprocessing.text import Tokenizer\n",
        "#from keras.preprocessing.sequence import pad_sequences\n",
        "from tqdm import tqdm\n",
        "from keras.models import Sequential\n",
        "from keras.layers import Embedding, LSTM, Dense, SpatialDropout1D, GRU\n",
        "from keras.callbacks import EarlyStopping\n",
        "from keras.initializers import Constant\n",
        "from sklearn.model_selection import train_test_split\n",
        "from tensorflow.keras.optimizers import Adam, Nadam, RMSprop\n",
        "from keras import Input\n",
        "\n",
        "from IPython.display import display"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "nltk.download('stopwords')\n",
        "nltk.download('wordnet')\n",
        "nltk.download('omw-1.4')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VGHly-zIpris",
        "outputId": "8a4ea7be-e80e-4a03-ee6f-d26e41d30236"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "[nltk_data] Downloading package stopwords to /root/nltk_data...\n",
            "[nltk_data]   Package stopwords is already up-to-date!\n",
            "[nltk_data] Downloading package wordnet to /root/nltk_data...\n",
            "[nltk_data]   Package wordnet is already up-to-date!\n",
            "[nltk_data] Downloading package omw-1.4 to /root/nltk_data...\n",
            "[nltk_data]   Package omw-1.4 is already up-to-date!\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "True"
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import warnings\n",
        "warnings.filterwarnings('ignore')"
      ],
      "metadata": {
        "id": "ovbC3RlUpxKS"
      },
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/gdrive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xqTk3PNGvHH2",
        "outputId": "b02688ed-2946-4834-c584-ca86ac9bdda3"
      },
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/gdrive; to attempt to forcibly remount, call drive.mount(\"/content/gdrive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df = pd.read_csv('/content/gdrive/MyDrive/yelp_dataset/yelp_reviews_clean_version4_st_louis.csv')"
      ],
      "metadata": {
        "id": "ofrho4sfu5nc"
      },
      "execution_count": 7,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "MfQeqg9yvQnR",
        "outputId": "377c01c2-3069-4db5-b2d6-b6045c0fa06e"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(292376, 5)"
            ]
          },
          "metadata": {},
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.rename(columns = {'text':'reviews'}, inplace = True)"
      ],
      "metadata": {
        "id": "hkBlD4FkBb9b"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ede7p75luxQy"
      },
      "source": [
        "**Extract year**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 10,
      "metadata": {
        "id": "SggCnX1zuxQy"
      },
      "outputs": [],
      "source": [
        "df['review_year'] = df['date'].apply(lambda x:int(x[:4]))"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Aspect-Based Sentiment Analysis"
      ],
      "metadata": {
        "id": "gL_wyDiH4ZNE"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "ABSA Text Preprocessing (SAMPLE)"
      ],
      "metadata": {
        "id": "GbhOroqN5c5o"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import missingno as msno\n",
        "plt.figure(figsize=(25, 20))\n",
        "msno.matrix(df, color=[0.2, 0.4, 1])\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 539
        },
        "id": "k1LAGZ6B4X9K",
        "outputId": "7b21563e-a9ee-4093-cd57-173c8d408a93"
      },
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1800x1440 with 0 Axes>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1800x720 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import re\n",
        "\n",
        "\n",
        "def remove_url(text):\n",
        "    text = re.sub(r\"http\\S+\", \"\", text)\n",
        "    return text\n",
        "\n",
        "\n",
        "# Example string with weird font characters plus an URL which we gonna remove.\n",
        "sample = \"This is a yelp project. \\n https://www.yelp.com 𝕒𝕟𝕕 𝕀 𝕒𝕞 😊 𝕥𝕠 𝕔𝕣𝕖𝕒𝕥𝕖 𝕥𝕙𝕚𝕤 𝕟𝕠𝕥𝕖𝕓𝕠𝕠𝕜\"\n",
        "print(f\"Text before removing url:- \\n {sample}\")\n",
        "\n",
        "sample = remove_url(sample)\n",
        "print(f\"Text after removing url:- \\n {sample}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Grrrha406KDC",
        "outputId": "f04cb6e5-1629-49c4-f12b-c4cf8e8e76f1"
      },
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Text before removing url:- \n",
            " This is a yelp project. \n",
            " https://www.yelp.com 𝕒𝕟𝕕 𝕀 𝕒𝕞 😊 𝕥𝕠 𝕔𝕣𝕖𝕒𝕥𝕖 𝕥𝕙𝕚𝕤 𝕟𝕠𝕥𝕖𝕓𝕠𝕠𝕜\n",
            "Text after removing url:- \n",
            " This is a yelp project. \n",
            "  𝕒𝕟𝕕 𝕀 𝕒𝕞 😊 𝕥𝕠 𝕔𝕣𝕖𝕒𝕥𝕖 𝕥𝕙𝕚𝕤 𝕟𝕠𝕥𝕖𝕓𝕠𝕠𝕜\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import unicodedata as uni\n",
        "\n",
        "print(f\"Text before Unicode Normalization:- \\n {sample}\")\n",
        "\n",
        "sample = uni.normalize('NFKD', sample)\n",
        "print(f\"Text after Unicode Normalization:- \\n {sample}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "6eshxmLV6DCe",
        "outputId": "69440aa1-d862-4607-e1f6-83d7e25289f9"
      },
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Text before Unicode Normalization:- \n",
            " This is a yelp project. \n",
            "  𝕒𝕟𝕕 𝕀 𝕒𝕞 😊 𝕥𝕠 𝕔𝕣𝕖𝕒𝕥𝕖 𝕥𝕙𝕚𝕤 𝕟𝕠𝕥𝕖𝕓𝕠𝕠𝕜\n",
            "Text after Unicode Normalization:- \n",
            " This is a yelp project. \n",
            "  and I am 😊 to create this notebook\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install demoji"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AmVtr6we5n5c",
        "outputId": "819e2ccd-c16e-48c8-8177-81a67f147ed8"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Requirement already satisfied: demoji in /usr/local/lib/python3.7/dist-packages (1.1.0)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import demoji\n",
        "\n",
        "\n",
        "def handle_emoji(string):\n",
        "    emojis = demoji.findall(string)\n",
        "\n",
        "    for emoji in emojis:\n",
        "        string = string.replace(emoji, \" \" + emojis[emoji].split(\":\")[0])\n",
        "\n",
        "    return string\n",
        "\n",
        "\n",
        "print(f\"Before Handling emoji:- \\n {sample}\")\n",
        "print(f\"After Handling emoji:- \\n {handle_emoji(sample)}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LzpulKJ05n2B",
        "outputId": "c1ba3994-1284-4385-b618-8c6e9aa160bb"
      },
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Before Handling emoji:- \n",
            " This is a yelp project. \n",
            "  and I am 😊 to create this notebook\n",
            "After Handling emoji:- \n",
            " This is a yelp project. \n",
            "  and I am  smiling face with smiling eyes to create this notebook\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def word_tokenizer(text):\n",
        "    text = text.lower()\n",
        "    text = text.split()\n",
        "\n",
        "    return text\n",
        "\n",
        "\n",
        "sample = \"This is a yelp project.\"\n",
        "print(sample)\n",
        "print(word_tokenizer(sample))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1UYUdMro5nx5",
        "outputId": "b06aafd3-3a00-4c9d-9d9b-49f370efd5b8"
      },
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "This is a yelp project.\n",
            "['this', 'is', 'a', 'yelp', 'project.']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from nltk.corpus import stopwords\n",
        "\n",
        "en_stopwords = set(stopwords.words('english'))\n",
        "print(f\"Stop Words in English : \\n{ en_stopwords}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C1JX5OhK5nvl",
        "outputId": "4472685b-5f63-40c2-d71b-18cfa3f41f6c"
      },
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Stop Words in English : \n",
            "{'their', 'few', 'i', 'don', 'to', 'has', 'too', 'down', \"you'll\", \"aren't\", \"that'll\", 'mightn', 'these', 'this', 'over', 'be', 's', \"wasn't\", 'him', 'should', 'll', 'your', \"weren't\", 'off', 'with', 'where', 'there', 'than', 'not', 'do', 'doesn', 'did', 'by', 'below', \"you'd\", 'yourselves', 'because', 'under', \"she's\", 'y', 'couldn', 'about', 'once', 'now', 'through', 'any', 'myself', 'when', 'but', 'ma', 'only', 'weren', \"it's\", 'other', 'yours', 'most', 'm', 'ourselves', 'in', 'above', 'such', 'nor', \"hadn't\", 't', \"mustn't\", 'been', 'into', 'our', 'before', \"don't\", 'on', 'can', 'a', 'won', 'needn', 'that', 'are', 'again', 'no', \"shan't\", 'further', \"isn't\", \"you're\", 'up', 'hasn', \"haven't\", 'wasn', 'his', 'itself', 'does', 'them', 'were', 'same', 'here', \"couldn't\", 'out', 'he', 'of', 'didn', 'you', 'shan', 'am', \"needn't\", 'shouldn', 'd', 'until', \"should've\", 'wouldn', 'having', 'some', 'had', \"won't\", 'as', 'all', \"shouldn't\", 'have', 'during', 'the', 'against', 'she', \"you've\", 'for', 'yourself', 'ours', \"hasn't\", 'or', 'so', 've', 'mustn', 'if', 'just', 'more', 'it', 'they', 'at', 'which', 'o', 'those', 'both', \"doesn't\", 'then', 'her', 're', 'after', 'how', 'isn', 'my', 'me', 'hers', 'whom', \"mightn't\", 'theirs', 'very', 'aren', 'we', 'doing', 'will', 'haven', 'what', \"didn't\", 'and', 'herself', 'himself', 'themselves', 'ain', 'is', 'was', 'an', 'being', 'hadn', 'own', 'each', 'while', 'who', \"wouldn't\", 'why', 'from', 'between', 'its'}\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "def remove_stopwords(text):\n",
        "    text = [word for word in text if word not in en_stopwords]\n",
        "    return text\n",
        "\n",
        "\n",
        "print(f\"Before removing stopwords : {word_tokenizer(sample)}\")\n",
        "print(f\"After removing stopwords : {remove_stopwords(word_tokenizer(sample))}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uSYC_KsC5ntK",
        "outputId": "4cb996c2-cf34-4941-9e9c-271d14d53b85"
      },
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Before removing stopwords : ['this', 'is', 'a', 'yelp', 'project.']\n",
            "After removing stopwords : ['yelp', 'project.']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from nltk.stem.porter import PorterStemmer\n",
        "from nltk.stem.lancaster import LancasterStemmer\n",
        "\n",
        "stemmer = PorterStemmer()\n",
        "# stemmer = LancasterStemmer()\n",
        "\n",
        "\n",
        "def stemming(text):\n",
        "\n",
        "    text = [stemmer.stem(word) for word in text]\n",
        "    return text\n",
        "\n",
        "\n",
        "sample = \"I am creating a Notebook\"\n",
        "print(f\"Before Stemming : {(sample)}\")\n",
        "print(f\"After Stemming : {stemming(word_tokenizer(sample))}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zdUn3mPW5nq5",
        "outputId": "3085ff65-7245-4dd5-b237-3e13d458f032"
      },
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Before Stemming : I am creating a Notebook\n",
            "After Stemming : ['i', 'am', 'creat', 'a', 'notebook']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "! pip install langdetect"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "J_pJ8-nEyVgs",
        "outputId": "e67c7d85-f860-4932-86b6-24717e9ad43e"
      },
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Looking in indexes: https://pypi.org/simple, https://us-python.pkg.dev/colab-wheels/public/simple/\n",
            "Requirement already satisfied: langdetect in /usr/local/lib/python3.7/dist-packages (1.0.9)\n",
            "Requirement already satisfied: six in /usr/local/lib/python3.7/dist-packages (from langdetect) (1.15.0)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from langdetect import detect\n",
        "\n",
        "en_text = u\"\"\"\"Hi Everyone I am Ankit Gupta.\"\"\"\n",
        "print(f\" {en_text} : {detect(en_text)}\")\n",
        "\n",
        "hindi_text = u\"\"\"मेरा नाम अंकित गुप्ता है और मैं एक छात्र हूँ\"\"\"\n",
        "print(f\"{hindi_text} : {detect(hindi_text)}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UQiBYBrJyMpi",
        "outputId": "7e52344b-7307-4e79-eedb-1fbe57ea1784"
      },
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " \"Hi Everyone I am Ankit Gupta. : tl\n",
            "मेरा नाम अंकित गुप्ता है और मैं एक छात्र हूँ : hi\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import spacy\n",
        "\n",
        "sp = spacy.load(\"en_core_web_sm\")"
      ],
      "metadata": {
        "id": "eQSqVYCk5noB"
      },
      "execution_count": 22,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def lemmatization(text):\n",
        "\n",
        "    # text = [sp(word).lemma_ for word in text]\n",
        "\n",
        "    text = \" \".join(text)\n",
        "    token = sp(text)\n",
        "\n",
        "    text = [word.lemma_ for word in token]\n",
        "    return text\n",
        "\n",
        "\n",
        "print(f\"Before Lemmatization : {word_tokenizer(sample)}\")\n",
        "print(f\"After Lemmatization : {lemmatization(word_tokenizer(sample))}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NsFoH26A5nlS",
        "outputId": "17cfa97f-6f6e-4f70-ba4f-04a7eb3d5913"
      },
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Before Lemmatization : ['i', 'am', 'creating', 'a', 'notebook']\n",
            "After Lemmatization : ['I', 'be', 'create', 'a', 'notebook']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp = df.copy()"
      ],
      "metadata": {
        "id": "eVUOnLER5nif"
      },
      "execution_count": 24,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "cMnf8tx561t8",
        "outputId": "770b0264-7600-490f-ec9f-67a4aa676d8b"
      },
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  stars  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01    4.0   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26    5.0   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11    4.0   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25    5.0   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32    4.0   \n",
              "\n",
              "                                             reviews  review_year  \n",
              "0  Love going here for happy hour or dinner!  Gre...         2014  \n",
              "1  My friends at work (connoisseurs of good food ...         2014  \n",
              "2  Great service, relaxing atmosphere and the foo...         2015  \n",
              "3  I went to Brio for the first time on Wednesday...         2015  \n",
              "4  I usually steer clear of the chain restaurants...         2016  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-2ef867a8-7690-4eda-877d-7e09b09dc8f4\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>stars</th>\n",
              "      <th>reviews</th>\n",
              "      <th>review_year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>2014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>5.0</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>2014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>2015</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>5.0</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>2015</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>4.0</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>2016</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-2ef867a8-7690-4eda-877d-7e09b09dc8f4')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-2ef867a8-7690-4eda-877d-7e09b09dc8f4 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-2ef867a8-7690-4eda-877d-7e09b09dc8f4');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp['stars'] = df['stars'].astype('int')"
      ],
      "metadata": {
        "id": "AfEVyuTp8MJ1"
      },
      "execution_count": 26,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "7OjMzCYx8RL1",
        "outputId": "6fd0a1dc-c92d-4033-c0b7-408b2a2794e4"
      },
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  stars  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01      4   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26      5   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11      4   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25      5   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32      4   \n",
              "\n",
              "                                             reviews  review_year  \n",
              "0  Love going here for happy hour or dinner!  Gre...         2014  \n",
              "1  My friends at work (connoisseurs of good food ...         2014  \n",
              "2  Great service, relaxing atmosphere and the foo...         2015  \n",
              "3  I went to Brio for the first time on Wednesday...         2015  \n",
              "4  I usually steer clear of the chain restaurants...         2016  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-3b898db1-09be-45ae-819a-ab9101d3a406\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>stars</th>\n",
              "      <th>reviews</th>\n",
              "      <th>review_year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>4</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>2014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>5</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>2014</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>4</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>2015</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>5</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>2015</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>4</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>2016</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3b898db1-09be-45ae-819a-ab9101d3a406')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-3b898db1-09be-45ae-819a-ab9101d3a406 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-3b898db1-09be-45ae-819a-ab9101d3a406');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "SWDt9D8wMQwH",
        "outputId": "53018e3a-56ee-4891-bc24-2f95b494bd9c"
      },
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               stars    review_year\n",
              "count  292376.000000  292376.000000\n",
              "mean        3.787113    2016.096964\n",
              "std         1.368116       3.196996\n",
              "min         1.000000    2005.000000\n",
              "25%         3.000000    2014.000000\n",
              "50%         4.000000    2016.000000\n",
              "75%         5.000000    2019.000000\n",
              "max         5.000000    2022.000000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1881fb80-a42f-40ae-b22e-8a5be59c4fd5\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>stars</th>\n",
              "      <th>review_year</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>3.787113</td>\n",
              "      <td>2016.096964</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>1.368116</td>\n",
              "      <td>3.196996</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>2005.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>3.000000</td>\n",
              "      <td>2014.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>4.000000</td>\n",
              "      <td>2016.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>5.000000</td>\n",
              "      <td>2019.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>5.000000</td>\n",
              "      <td>2022.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1881fb80-a42f-40ae-b22e-8a5be59c4fd5')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1881fb80-a42f-40ae-b22e-8a5be59c4fd5 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1881fb80-a42f-40ae-b22e-8a5be59c4fd5');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Assigning labels to each review\n",
        "\n",
        "def label(stars):\n",
        "    if stars == 5:\n",
        "        return 1\n",
        "    elif stars == 4:\n",
        "        return 1\n",
        "    else:\n",
        "        return 0\n",
        "    \n",
        "from tqdm import tqdm\n",
        "\n",
        "tqdm.pandas()\n",
        "\n",
        "df_temp['y'] = df_temp.stars.progress_map(label)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "zG6ulFy25ndX",
        "outputId": "0f7bba0f-0752-4602-ad25-d7fdad5fd5da"
      },
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 292376/292376 [00:00<00:00, 749332.79it/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp = df_temp[['user_id', 'business_id', 'date', 'review_year', 'reviews', 'y', 'stars']]"
      ],
      "metadata": {
        "id": "vcIUiWCR5nXZ"
      },
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "cIIp1XwZ5nQ9",
        "outputId": "a4476cfd-f322-4320-94b8-abcb83132158"
      },
      "execution_count": 31,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32   \n",
              "\n",
              "   review_year                                            reviews  y  stars  \n",
              "0         2014  Love going here for happy hour or dinner!  Gre...  1      4  \n",
              "1         2014  My friends at work (connoisseurs of good food ...  1      5  \n",
              "2         2015  Great service, relaxing atmosphere and the foo...  1      4  \n",
              "3         2015  I went to Brio for the first time on Wednesday...  1      5  \n",
              "4         2016  I usually steer clear of the chain restaurants...  1      4  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-0ecfd76b-5032-4382-816c-3f8578a4d679\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>review_year</th>\n",
              "      <th>reviews</th>\n",
              "      <th>y</th>\n",
              "      <th>stars</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>2014</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>2014</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>2015</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>2015</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>2016</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0ecfd76b-5032-4382-816c-3f8578a4d679')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-0ecfd76b-5032-4382-816c-3f8578a4d679 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-0ecfd76b-5032-4382-816c-3f8578a4d679');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 31
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp.describe()"
      ],
      "metadata": {
        "id": "qL0liLPQ97il",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "outputId": "a4ecb337-1c1f-477f-f33e-8047fce03416"
      },
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "         review_year              y          stars\n",
              "count  292376.000000  292376.000000  292376.000000\n",
              "mean     2016.096964       0.676851       3.787113\n",
              "std         3.196996       0.467680       1.368116\n",
              "min      2005.000000       0.000000       1.000000\n",
              "25%      2014.000000       0.000000       3.000000\n",
              "50%      2016.000000       1.000000       4.000000\n",
              "75%      2019.000000       1.000000       5.000000\n",
              "max      2022.000000       1.000000       5.000000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-90d4d298-3fbb-467e-9581-de8495f6a8e7\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>review_year</th>\n",
              "      <th>y</th>\n",
              "      <th>stars</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>2016.096964</td>\n",
              "      <td>0.676851</td>\n",
              "      <td>3.787113</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>3.196996</td>\n",
              "      <td>0.467680</td>\n",
              "      <td>1.368116</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>2005.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>2014.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>3.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>2016.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>4.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>2019.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>5.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>2022.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>5.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-90d4d298-3fbb-467e-9581-de8495f6a8e7')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-90d4d298-3fbb-467e-9581-de8495f6a8e7 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-90d4d298-3fbb-467e-9581-de8495f6a8e7');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp.y.value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LgU9MAck5nGE",
        "outputId": "ee66dff8-2a33-4ce6-9c43-9eb8b15d505d"
      },
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1    197895\n",
              "0     94481\n",
              "Name: y, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 33
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "\n",
        "sns.countplot(x = \"y\", data = df_temp)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "OkEIY2KN5ltd",
        "outputId": "ca5207f7-a937-439e-b05d-abec74ccbeae"
      },
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp2 = df_temp[(df_temp['stars'] == 5)]"
      ],
      "metadata": {
        "id": "U7Vht4gW-ULc"
      },
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp2.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "83ko9DfDPJ6s",
        "outputId": "7875e41b-c71a-47c7-bc50-7e943b67ef8f"
      },
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "Int64Index: 124247 entries, 1 to 292375\n",
            "Data columns (total 7 columns):\n",
            " #   Column       Non-Null Count   Dtype \n",
            "---  ------       --------------   ----- \n",
            " 0   user_id      124247 non-null  object\n",
            " 1   business_id  124247 non-null  object\n",
            " 2   date         124247 non-null  object\n",
            " 3   review_year  124247 non-null  int64 \n",
            " 4   reviews      124247 non-null  object\n",
            " 5   y            124247 non-null  int64 \n",
            " 6   stars        124247 non-null  int64 \n",
            "dtypes: int64(3), object(4)\n",
            "memory usage: 7.6+ MB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_temp2.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "bPy2llAZUw45",
        "outputId": "87f4280e-d2b2-4f56-9776-af340ea5ea12"
      },
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                   user_id             business_id                 date  \\\n",
              "1   5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26   \n",
              "3   st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25   \n",
              "6   kX9OMmA9XLpNPWXQtTe33Q  lj-E32x9_FA7GmUrBGBEWg  2017-02-09 19:39:40   \n",
              "16  gjzKWzc8ZPn5Aylf8TMfTg  lj-E32x9_FA7GmUrBGBEWg  2015-11-21 16:11:54   \n",
              "25  4xV0D0udUpKbm5k7MxhfnA  lj-E32x9_FA7GmUrBGBEWg  2018-07-29 02:41:42   \n",
              "\n",
              "    review_year                                            reviews  y  stars  \n",
              "1          2014  My friends at work (connoisseurs of good food ...  1      5  \n",
              "3          2015  I went to Brio for the first time on Wednesday...  1      5  \n",
              "6          2017  I love this cute Italian restaurant, it's grea...  1      5  \n",
              "16         2015  I've been here 2 times now. The food is absolu...  1      5  \n",
              "25         2018  Amazing anniversary dinner. Staff all wished u...  1      5  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-7ff39680-0318-499c-9fb4-495e52f3c12a\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>review_year</th>\n",
              "      <th>reviews</th>\n",
              "      <th>y</th>\n",
              "      <th>stars</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>2014</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>2015</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>kX9OMmA9XLpNPWXQtTe33Q</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2017-02-09 19:39:40</td>\n",
              "      <td>2017</td>\n",
              "      <td>I love this cute Italian restaurant, it's grea...</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>gjzKWzc8ZPn5Aylf8TMfTg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-11-21 16:11:54</td>\n",
              "      <td>2015</td>\n",
              "      <td>I've been here 2 times now. The food is absolu...</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>4xV0D0udUpKbm5k7MxhfnA</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2018-07-29 02:41:42</td>\n",
              "      <td>2018</td>\n",
              "      <td>Amazing anniversary dinner. Staff all wished u...</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-7ff39680-0318-499c-9fb4-495e52f3c12a')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-7ff39680-0318-499c-9fb4-495e52f3c12a button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-7ff39680-0318-499c-9fb4-495e52f3c12a');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "ABSA Text Preprocessing (ENTIRE TEXT)"
      ],
      "metadata": {
        "id": "-ayxYuNF_q2A"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def preprocessing(text):\n",
        "    \n",
        "    text = remove_url(text) \n",
        "    text = uni.normalize('NFKD', text)\n",
        "    text = handle_emoji(text)\n",
        "    text = text.lower() \n",
        "    text = re.sub(r'[^\\w\\s]', '', text)\n",
        "    text = word_tokenizer(text)\n",
        "    #text = stemming(text)\n",
        "    text = lemmatization(text)\n",
        "    text = remove_stopwords(text)\n",
        "    text = \" \".join(text)\n",
        "\n",
        "    return text"
      ],
      "metadata": {
        "id": "IkF5ZgM7_c7U"
      },
      "execution_count": 38,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tqdm import tqdm\n",
        "\n",
        "tqdm.pandas()\n",
        "\n",
        "df['clean_review'] = df['reviews'].progress_map(preprocessing)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "G1CmVCof_c4s",
        "outputId": "3da6cd70-3b8b-4083-da6a-f9c7669cdf5d"
      },
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 292376/292376 [2:14:35<00:00, 36.20it/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 337
        },
        "id": "J2fQXmRC_cs9",
        "outputId": "891677aa-50f9-44a8-a0a2-d6c1dd8a38b2"
      },
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  stars  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01    4.0   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26    5.0   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11    4.0   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25    5.0   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32    4.0   \n",
              "\n",
              "                                             reviews  review_year  \\\n",
              "0  Love going here for happy hour or dinner!  Gre...         2014   \n",
              "1  My friends at work (connoisseurs of good food ...         2014   \n",
              "2  Great service, relaxing atmosphere and the foo...         2015   \n",
              "3  I went to Brio for the first time on Wednesday...         2015   \n",
              "4  I usually steer clear of the chain restaurants...         2016   \n",
              "\n",
              "                                        clean_review  \n",
              "0  love go happy hour dinner great patio fan beat...  \n",
              "1  friend work connoisseur good food st louis are...  \n",
              "2  great service relax atmosphere food exceptiona...  \n",
              "3  I go brio first time wednesday dinner friend s...  \n",
              "4  I usually steer clear chain restaurant since m...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a75208bb-3978-480f-97bf-dfd560dab20c\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>stars</th>\n",
              "      <th>reviews</th>\n",
              "      <th>review_year</th>\n",
              "      <th>clean_review</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>2014</td>\n",
              "      <td>love go happy hour dinner great patio fan beat...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>5.0</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>2014</td>\n",
              "      <td>friend work connoisseur good food st louis are...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>2015</td>\n",
              "      <td>great service relax atmosphere food exceptiona...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>5.0</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>2015</td>\n",
              "      <td>I go brio first time wednesday dinner friend s...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>4.0</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>2016</td>\n",
              "      <td>I usually steer clear chain restaurant since m...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a75208bb-3978-480f-97bf-dfd560dab20c')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-a75208bb-3978-480f-97bf-dfd560dab20c button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-a75208bb-3978-480f-97bf-dfd560dab20c');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "reviews = df.clean_review.values.tolist()"
      ],
      "metadata": {
        "id": "g52ZH_ZnAZbP"
      },
      "execution_count": 41,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tqdm import tqdm\n",
        "\n",
        "tqdm.pandas()\n",
        "\n",
        "df['clean_review2'] = df['clean_review'].progress_map(word_tokenizer)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "W3o8YZyDAZY9",
        "outputId": "e48e4fa8-c153-4de3-9c05-0abb2d16097f"
      },
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 292376/292376 [00:05<00:00, 58459.25it/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 424
        },
        "id": "9NtQqxlmAZWw",
        "outputId": "67811844-42bc-44d3-b5aa-78dd6faa2faa"
      },
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  stars  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01    4.0   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26    5.0   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11    4.0   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25    5.0   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32    4.0   \n",
              "\n",
              "                                             reviews  review_year  \\\n",
              "0  Love going here for happy hour or dinner!  Gre...         2014   \n",
              "1  My friends at work (connoisseurs of good food ...         2014   \n",
              "2  Great service, relaxing atmosphere and the foo...         2015   \n",
              "3  I went to Brio for the first time on Wednesday...         2015   \n",
              "4  I usually steer clear of the chain restaurants...         2016   \n",
              "\n",
              "                                        clean_review  \\\n",
              "0  love go happy hour dinner great patio fan beat...   \n",
              "1  friend work connoisseur good food st louis are...   \n",
              "2  great service relax atmosphere food exceptiona...   \n",
              "3  I go brio first time wednesday dinner friend s...   \n",
              "4  I usually steer clear chain restaurant since m...   \n",
              "\n",
              "                                       clean_review2  \n",
              "0  [love, go, happy, hour, dinner, great, patio, ...  \n",
              "1  [friend, work, connoisseur, good, food, st, lo...  \n",
              "2  [great, service, relax, atmosphere, food, exce...  \n",
              "3  [i, go, brio, first, time, wednesday, dinner, ...  \n",
              "4  [i, usually, steer, clear, chain, restaurant, ...  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-111c1a58-3cc5-46de-bf40-ddbea3b9e35c\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>stars</th>\n",
              "      <th>reviews</th>\n",
              "      <th>review_year</th>\n",
              "      <th>clean_review</th>\n",
              "      <th>clean_review2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>2014</td>\n",
              "      <td>love go happy hour dinner great patio fan beat...</td>\n",
              "      <td>[love, go, happy, hour, dinner, great, patio, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>5.0</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>2014</td>\n",
              "      <td>friend work connoisseur good food st louis are...</td>\n",
              "      <td>[friend, work, connoisseur, good, food, st, lo...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>2015</td>\n",
              "      <td>great service relax atmosphere food exceptiona...</td>\n",
              "      <td>[great, service, relax, atmosphere, food, exce...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>5.0</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>2015</td>\n",
              "      <td>I go brio first time wednesday dinner friend s...</td>\n",
              "      <td>[i, go, brio, first, time, wednesday, dinner, ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>4.0</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>2016</td>\n",
              "      <td>I usually steer clear chain restaurant since m...</td>\n",
              "      <td>[i, usually, steer, clear, chain, restaurant, ...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-111c1a58-3cc5-46de-bf40-ddbea3b9e35c')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-111c1a58-3cc5-46de-bf40-ddbea3b9e35c button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-111c1a58-3cc5-46de-bf40-ddbea3b9e35c');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "data_words = df['clean_review2'].values.tolist()\n",
        "len(data_words)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IgCSVFvpAZUY",
        "outputId": "03028441-fb0e-4e00-946c-2c6987966175"
      },
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "292376"
            ]
          },
          "metadata": {},
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#save checkpoint\n",
        "df.to_csv(\"/content/gdrive/MyDrive/yelp_dataset/Clean_Yelp_ABSA_Reviews2.csv\", index = False)"
      ],
      "metadata": {
        "id": "QCSOAfXXdi8T"
      },
      "execution_count": 45,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "import gensim.corpora as corpora\n",
        "\n",
        "# Create Dictionary\n",
        "id2word = corpora.Dictionary(data_words)\n",
        "# Create Corpus\n",
        "texts = data_words\n",
        "# Term Document Frequency\n",
        "corpus = [id2word.doc2bow(text) for text in texts]\n",
        "# View\n",
        "print(corpus[:1][0][:30])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hh6nW1izAZRM",
        "outputId": "155d85d5-2108-41ff-e267-746ed2ce0ef6"
      },
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[(0, 1), (1, 1), (2, 1), (3, 1), (4, 1), (5, 1), (6, 1), (7, 1), (8, 1), (9, 1), (10, 1), (11, 1), (12, 1), (13, 1), (14, 1), (15, 1), (16, 1), (17, 1), (18, 1), (19, 1), (20, 1), (21, 1), (22, 1), (23, 1), (24, 1), (25, 1)]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from gensim.models import LdaMulticore\n",
        "from gensim.models import LdaModel\n",
        "from pprint import pprint\n",
        "\n",
        "# number of topics\n",
        "num_topics = 10\n",
        "# Build LDA model\n",
        "lda_model = LdaMulticore(corpus=corpus, id2word=id2word,\n",
        "                     num_topics=num_topics, iterations=400)\n",
        "# Print the Keyword in the 10 topics\n",
        "pprint(lda_model.print_topics())\n",
        "doc_lda = lda_model[corpus]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qbBJMdwzAZHy",
        "outputId": "4a75ef98-657a-456c-803d-5ba5daadd4d7"
      },
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[(0,\n",
            "  '0.105*\"pizza\" + 0.033*\"salad\" + 0.024*\"i\" + 0.018*\"cheese\" + 0.016*\"crust\" '\n",
            "  '+ 0.013*\"order\" + 0.013*\"good\" + 0.010*\"sauce\" + 0.010*\"topping\" + '\n",
            "  '0.009*\"get\"'),\n",
            " (1,\n",
            "  '0.056*\"i\" + 0.029*\"good\" + 0.026*\"food\" + 0.026*\"st\" + 0.024*\"louis\" + '\n",
            "  '0.022*\"place\" + 0.017*\"restaurant\" + 0.013*\"try\" + 0.013*\"thai\" + '\n",
            "  '0.011*\"dish\"'),\n",
            " (2,\n",
            "  '0.115*\"i\" + 0.018*\"order\" + 0.016*\"food\" + 0.014*\"go\" + 0.013*\"get\" + '\n",
            "  '0.011*\"time\" + 0.010*\"say\" + 0.010*\"place\" + 0.009*\"would\" + 0.009*\"like\"'),\n",
            " (3,\n",
            "  '0.026*\"wait\" + 0.021*\"table\" + 0.019*\"food\" + 0.019*\"get\" + 0.019*\"minute\" '\n",
            "  '+ 0.017*\"come\" + 0.017*\"order\" + 0.015*\"drink\" + 0.015*\"take\" + '\n",
            "  '0.014*\"time\"'),\n",
            " (4,\n",
            "  '0.044*\"chicken\" + 0.039*\"fry\" + 0.026*\"sandwich\" + 0.022*\"burger\" + '\n",
            "  '0.019*\"good\" + 0.018*\"cheese\" + 0.017*\"sauce\" + 0.013*\"order\" + '\n",
            "  '0.013*\"potato\" + 0.012*\"side\"'),\n",
            " (5,\n",
            "  '0.015*\"wine\" + 0.014*\"steak\" + 0.013*\"dinner\" + 0.012*\"restaurant\" + '\n",
            "  '0.010*\"experience\" + 0.009*\"night\" + 0.009*\"dining\" + 0.009*\"room\" + '\n",
            "  '0.009*\"great\" + 0.009*\"make\"'),\n",
            " (6,\n",
            "  '0.065*\"i\" + 0.019*\"good\" + 0.011*\"order\" + 0.011*\"get\" + 0.010*\"roll\" + '\n",
            "  '0.009*\"sushi\" + 0.009*\"like\" + 0.009*\"try\" + 0.008*\"come\" + 0.008*\"also\"'),\n",
            " (7,\n",
            "  '0.047*\"chip\" + 0.047*\"taco\" + 0.032*\"fish\" + 0.029*\"salsa\" + 0.022*\"nachos\" '\n",
            "  '+ 0.015*\"rice\" + 0.014*\"good\" + 0.014*\"korean\" + 0.011*\"bean\" + '\n",
            "  '0.010*\"raman\"'),\n",
            " (8,\n",
            "  '0.059*\"i\" + 0.013*\"good\" + 0.012*\"like\" + 0.010*\"place\" + 0.009*\"get\" + '\n",
            "  '0.007*\"bbq\" + 0.007*\"well\" + 0.006*\"go\" + 0.006*\"one\" + 0.006*\"beer\"'),\n",
            " (9,\n",
            "  '0.052*\"great\" + 0.044*\"food\" + 0.035*\"good\" + 0.032*\"place\" + 0.031*\"i\" + '\n",
            "  '0.026*\"service\" + 0.016*\"love\" + 0.016*\"friendly\" + 0.016*\"go\" + '\n",
            "  '0.015*\"always\"')]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#data_words = df['clean_review2']\n",
        "\n",
        "#def gen_words(data_words):\n",
        "#    yield data_words\n",
        "data_words = df['clean_review2'].values.tolist()\n",
        "len(data_words)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "oyreQs_2Va9V",
        "outputId": "4ff12f08-f8b0-419c-9097-90877114b622"
      },
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "292376"
            ]
          },
          "metadata": {},
          "execution_count": 48
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "%%time\n",
        "from gensim.models import FastText\n",
        "fasttext_model = FastText(data_words, size=100, window=5, min_count=5, workers=4,sg=1)\n",
        "# fasttext_model = FastText.load_fasttext_format(\"../input/fast100/cc.en.100.bin\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Tk343-sPAoAC",
        "outputId": "57270285-a725-4bd5-9b59-8730451791b4"
      },
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "CPU times: user 25min 47s, sys: 3.36 s, total: 25min 50s\n",
            "Wall time: 13min 26s\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fasttext_model.save(\"FastText-Model-For-ABSA.bin\")"
      ],
      "metadata": {
        "id": "wyZW3SamAnsv"
      },
      "execution_count": 50,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(fasttext_model.wv['service'])"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vUKd02e7g1zb",
        "outputId": "ffd24dc0-df0e-46bf-95d7-c6f42a0fb5b6"
      },
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "[ 0.11161157  0.13919845 -0.1320141  -0.13125886  0.24426009 -0.0766753\n",
            " -0.03265483 -0.2534362   0.09373141 -0.25906444  0.1756083   0.06498928\n",
            "  0.03394561  0.00783977 -0.41098693  0.0187496  -0.39475632  0.20405386\n",
            " -0.26476985 -0.15422079  0.32190934  0.31294486 -0.450854    0.3762911\n",
            "  0.51335096 -0.05387257 -0.12655723  0.20022312  0.21737675 -0.16788492\n",
            " -0.22382095 -0.1741892   0.27145514 -0.02698541  0.5913705  -0.36435422\n",
            "  0.07780357  0.01718424  0.3165588   0.04981562  0.34859073 -0.3621168\n",
            " -0.51721543 -0.32500473 -0.4370381  -0.2893733   0.23029533 -0.11282208\n",
            " -0.15471359 -0.1959459  -0.2926862   0.25445896  0.11593823 -0.13931152\n",
            "  0.22482131  0.07110508  0.12855722 -0.06902418 -0.19909674  0.28235316\n",
            "  0.02575272  0.17689599  0.39664373  0.08227926 -0.19386992  0.41935635\n",
            " -0.09664096 -0.26038542 -0.08585173 -0.1474843   0.06199944 -0.16014521\n",
            " -0.4099759  -0.2465112  -0.05851781 -0.473936    0.17616935 -0.18061118\n",
            "  0.01961164 -0.11829029  0.27819687 -0.05283042  0.16889639 -0.07429101\n",
            " -0.02256591  0.18648513 -0.00713282  0.0570677  -0.18152688  0.1917108\n",
            " -0.1296426  -0.14047177 -0.45942158 -0.27275372 -0.0084429  -0.46212167\n",
            " -0.2720086  -0.31595206 -0.00571382  0.22388856]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "semantically_similar_words = {words: [item[0] for item in fasttext_model.wv.most_similar([words], topn=5)]\n",
        "                  for words in ['food', 'service', 'place', 'price']}\n",
        "\n",
        "for k,v in semantically_similar_words.items():\n",
        "    print(k+\":\"+str(v))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ea4Hfz8ch5XL",
        "outputId": "e21f88b1-a2d8-458a-e5da-6607ab6fb404"
      },
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "food:['food3', 'food4', 'fooda', 'foodall', 'foodoh']\n",
            "service:['serviceso', 'service4', 'serviceit', 'serviced', 'serviceour']\n",
            "place:['placea', 'placeit', 'placeif', 'placefood', 'placeso']\n",
            "price:['pricing', 'priced', 'pricei', 'pricetag', 'pricer']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(fasttext_model.wv.similarity(w1='food', w2='service'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CQip-CnZiGHc",
        "outputId": "50095d15-0ca2-4f3e-a747-3d62f7811196"
      },
      "execution_count": 53,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.59191245\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(fasttext_model.wv.similarity(w1='food', w2='place'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LoYw1ZShcZin",
        "outputId": "380c37c1-2342-4f00-80ca-aec4f4762e03"
      },
      "execution_count": 54,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.5579984\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(fasttext_model.wv.similarity(w1='food', w2='price'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "v6c8wWQ-ccY8",
        "outputId": "b6135fa6-a5b5-4d78-9d4c-47eea7098a52"
      },
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.47630614\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(fasttext_model.wv.similarity(w1='place', w2='service'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wI9Pyh4rcgi5",
        "outputId": "46ac51f8-6c73-4a87-af09-50f683a24ef9"
      },
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.46273276\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(fasttext_model.wv.similarity(w1='place', w2='price'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HZNg9mPNczW8",
        "outputId": "b64d3243-1161-4f12-c27f-2f4114cf5940"
      },
      "execution_count": 57,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.4090378\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "print(fasttext_model.wv.similarity(w1='service', w2='price'))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wIb_2kfmEcB0",
        "outputId": "18350559-f1da-4b6f-8593-828662f3d18d"
      },
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "0.4894688\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fasttext_model.wv.similarity('I was very surprised by this place', 'place')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3Q1HenrnAnqW",
        "outputId": "7970e967-9f41-4a4f-ab70-675e7da5155e"
      },
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.66795117"
            ]
          },
          "metadata": {},
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fasttext_model.wv.similarity(\"I was very surprised by this place\", \"service\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ScBG5nRQAnoN",
        "outputId": "8c341b12-4396-4f43-9a41-a1de74e6bfb8"
      },
      "execution_count": 60,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.47162268"
            ]
          },
          "metadata": {},
          "execution_count": 60
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fasttext_model.wv.similarity(\"I was very surprised by this place\", \"food\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SGWrApQcElL3",
        "outputId": "371e5792-9abd-417a-cb68-fdcf00cac10f"
      },
      "execution_count": 61,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.59318775"
            ]
          },
          "metadata": {},
          "execution_count": 61
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fasttext_model.wv.similarity(\"I was very surprised by this place\", \"price\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T1Pj4rgsEk_B",
        "outputId": "21ff0e6f-4978-4a18-baea-a1fc8bf7992b"
      },
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.4741543"
            ]
          },
          "metadata": {},
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "aspects = [\"food\", \"service\", \"place\", \"price\"]\n",
        "\n",
        "def get_similarity(text, aspect):\n",
        "    try:\n",
        "        text = \" \".join(text)\n",
        "        return fasttext_model.wv.similarity(text, aspect)\n",
        "    except:\n",
        "        return 0"
      ],
      "metadata": {
        "id": "CqaaGNyNAnmM"
      },
      "execution_count": 63,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tqdm import tqdm\n",
        "tqdm.pandas()\n",
        "for aspect in aspects:\n",
        "    df[aspect] = df['clean_review2'].progress_map(lambda text: get_similarity(text, aspect))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Y02X3TunAnjr",
        "outputId": "5aa7aa21-b583-44d3-94e1-b45d7a7a0ce7"
      },
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 292376/292376 [08:47<00:00, 553.89it/s]\n",
            "100%|██████████| 292376/292376 [08:42<00:00, 559.55it/s]\n",
            "100%|██████████| 292376/292376 [08:48<00:00, 553.70it/s]\n",
            "100%|██████████| 292376/292376 [08:45<00:00, 556.32it/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 565
        },
        "id": "znh_asmTXWjY",
        "outputId": "13ac1a35-249d-4a00-e628-709cf1e10939"
      },
      "execution_count": 81,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  stars  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01    4.0   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26    5.0   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11    4.0   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25    5.0   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32    4.0   \n",
              "\n",
              "                                             reviews  review_year  \\\n",
              "0  Love going here for happy hour or dinner!  Gre...         2014   \n",
              "1  My friends at work (connoisseurs of good food ...         2014   \n",
              "2  Great service, relaxing atmosphere and the foo...         2015   \n",
              "3  I went to Brio for the first time on Wednesday...         2015   \n",
              "4  I usually steer clear of the chain restaurants...         2016   \n",
              "\n",
              "                                        clean_review  \\\n",
              "0  love go happy hour dinner great patio fan beat...   \n",
              "1  friend work connoisseur good food st louis are...   \n",
              "2  great service relax atmosphere food exceptiona...   \n",
              "3  I go brio first time wednesday dinner friend s...   \n",
              "4  I usually steer clear chain restaurant since m...   \n",
              "\n",
              "                                       clean_review2      food   service  \\\n",
              "0  [love, go, happy, hour, dinner, great, patio, ...  0.533379  0.550373   \n",
              "1  [friend, work, connoisseur, good, food, st, lo...  0.603309  0.578850   \n",
              "2  [great, service, relax, atmosphere, food, exce...  0.635701  0.723729   \n",
              "3  [i, go, brio, first, time, wednesday, dinner, ...  0.614665  0.597369   \n",
              "4  [i, usually, steer, clear, chain, restaurant, ...  0.565750  0.524968   \n",
              "\n",
              "      place     price  \n",
              "0  0.604214  0.501414  \n",
              "1  0.592573  0.506964  \n",
              "2  0.620446  0.501792  \n",
              "3  0.633347  0.536175  \n",
              "4  0.690019  0.501281  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-fcf9baed-9186-47cc-9d86-f020fd46750c\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>stars</th>\n",
              "      <th>reviews</th>\n",
              "      <th>review_year</th>\n",
              "      <th>clean_review</th>\n",
              "      <th>clean_review2</th>\n",
              "      <th>food</th>\n",
              "      <th>service</th>\n",
              "      <th>place</th>\n",
              "      <th>price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>2014</td>\n",
              "      <td>love go happy hour dinner great patio fan beat...</td>\n",
              "      <td>[love, go, happy, hour, dinner, great, patio, ...</td>\n",
              "      <td>0.533379</td>\n",
              "      <td>0.550373</td>\n",
              "      <td>0.604214</td>\n",
              "      <td>0.501414</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>5.0</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>2014</td>\n",
              "      <td>friend work connoisseur good food st louis are...</td>\n",
              "      <td>[friend, work, connoisseur, good, food, st, lo...</td>\n",
              "      <td>0.603309</td>\n",
              "      <td>0.578850</td>\n",
              "      <td>0.592573</td>\n",
              "      <td>0.506964</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>2015</td>\n",
              "      <td>great service relax atmosphere food exceptiona...</td>\n",
              "      <td>[great, service, relax, atmosphere, food, exce...</td>\n",
              "      <td>0.635701</td>\n",
              "      <td>0.723729</td>\n",
              "      <td>0.620446</td>\n",
              "      <td>0.501792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>5.0</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>2015</td>\n",
              "      <td>I go brio first time wednesday dinner friend s...</td>\n",
              "      <td>[i, go, brio, first, time, wednesday, dinner, ...</td>\n",
              "      <td>0.614665</td>\n",
              "      <td>0.597369</td>\n",
              "      <td>0.633347</td>\n",
              "      <td>0.536175</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>4.0</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>2016</td>\n",
              "      <td>I usually steer clear chain restaurant since m...</td>\n",
              "      <td>[i, usually, steer, clear, chain, restaurant, ...</td>\n",
              "      <td>0.565750</td>\n",
              "      <td>0.524968</td>\n",
              "      <td>0.690019</td>\n",
              "      <td>0.501281</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-fcf9baed-9186-47cc-9d86-f020fd46750c')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-fcf9baed-9186-47cc-9d86-f020fd46750c button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-fcf9baed-9186-47cc-9d86-f020fd46750c');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 81
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#save checkpoint\n",
        "df.to_csv(\"/content/gdrive/MyDrive/yelp_dataset/Clean_Yelp_ABSA_Reviews2.csv\", index = False)"
      ],
      "metadata": {
        "id": "k0y7u9gjO4W-"
      },
      "execution_count": 82,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "SvOfKOq7Gqnp",
        "outputId": "705b9776-9309-49be-abd7-57af8fcb3a25"
      },
      "execution_count": 83,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               stars    review_year           food        service  \\\n",
              "count  292376.000000  292376.000000  292376.000000  292376.000000   \n",
              "mean        3.787113    2016.096964       0.577500       0.544896   \n",
              "std         1.368116       3.196996       0.047507       0.067984   \n",
              "min         1.000000    2005.000000       0.000000      -0.046627   \n",
              "25%         3.000000    2014.000000       0.549304       0.500160   \n",
              "50%         4.000000    2016.000000       0.580594       0.542505   \n",
              "75%         5.000000    2019.000000       0.609243       0.587919   \n",
              "max         5.000000    2022.000000       0.768563       0.846891   \n",
              "\n",
              "               place          price  \n",
              "count  292376.000000  292376.000000  \n",
              "mean        0.596461       0.508878  \n",
              "std         0.051792       0.048489  \n",
              "min         0.000000       0.000000  \n",
              "25%         0.564553       0.480152  \n",
              "50%         0.599314       0.508434  \n",
              "75%         0.631412       0.536732  \n",
              "max         0.817265       0.778234  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a57f5769-6302-4701-847e-a1533f9b4b6b\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>stars</th>\n",
              "      <th>review_year</th>\n",
              "      <th>food</th>\n",
              "      <th>service</th>\n",
              "      <th>place</th>\n",
              "      <th>price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>3.787113</td>\n",
              "      <td>2016.096964</td>\n",
              "      <td>0.577500</td>\n",
              "      <td>0.544896</td>\n",
              "      <td>0.596461</td>\n",
              "      <td>0.508878</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>1.368116</td>\n",
              "      <td>3.196996</td>\n",
              "      <td>0.047507</td>\n",
              "      <td>0.067984</td>\n",
              "      <td>0.051792</td>\n",
              "      <td>0.048489</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>2005.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>-0.046627</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>3.000000</td>\n",
              "      <td>2014.000000</td>\n",
              "      <td>0.549304</td>\n",
              "      <td>0.500160</td>\n",
              "      <td>0.564553</td>\n",
              "      <td>0.480152</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>4.000000</td>\n",
              "      <td>2016.000000</td>\n",
              "      <td>0.580594</td>\n",
              "      <td>0.542505</td>\n",
              "      <td>0.599314</td>\n",
              "      <td>0.508434</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>5.000000</td>\n",
              "      <td>2019.000000</td>\n",
              "      <td>0.609243</td>\n",
              "      <td>0.587919</td>\n",
              "      <td>0.631412</td>\n",
              "      <td>0.536732</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>5.000000</td>\n",
              "      <td>2022.000000</td>\n",
              "      <td>0.768563</td>\n",
              "      <td>0.846891</td>\n",
              "      <td>0.817265</td>\n",
              "      <td>0.778234</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a57f5769-6302-4701-847e-a1533f9b4b6b')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-a57f5769-6302-4701-847e-a1533f9b4b6b button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-a57f5769-6302-4701-847e-a1533f9b4b6b');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 83
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.dtypes"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "n36rpHrNNQ3i",
        "outputId": "7ffe2d3a-9e3d-4ee1-b39c-0f79d13f4519"
      },
      "execution_count": 84,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "user_id           object\n",
              "business_id       object\n",
              "date              object\n",
              "stars            float64\n",
              "reviews           object\n",
              "review_year        int64\n",
              "clean_review      object\n",
              "clean_review2     object\n",
              "food             float64\n",
              "service          float64\n",
              "place            float64\n",
              "price            float64\n",
              "dtype: object"
            ]
          },
          "metadata": {},
          "execution_count": 84
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#rename columns\n",
        "df.rename(columns = {'food':'food_similarity'}, inplace = True)\n",
        "df.rename(columns = {'service':'service_similarity'}, inplace = True)\n",
        "df.rename(columns = {'place':'place_similarity'}, inplace = True)\n",
        "df.rename(columns = {'price':'price_similarity'}, inplace = True)"
      ],
      "metadata": {
        "id": "HY9PZDhdFsUE"
      },
      "execution_count": 85,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 565
        },
        "id": "zeZs0_v5Og62",
        "outputId": "3975842b-7412-4eb8-b80f-1c0a5d99df16"
      },
      "execution_count": 86,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  stars  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01    4.0   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26    5.0   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11    4.0   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25    5.0   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32    4.0   \n",
              "\n",
              "                                             reviews  review_year  \\\n",
              "0  Love going here for happy hour or dinner!  Gre...         2014   \n",
              "1  My friends at work (connoisseurs of good food ...         2014   \n",
              "2  Great service, relaxing atmosphere and the foo...         2015   \n",
              "3  I went to Brio for the first time on Wednesday...         2015   \n",
              "4  I usually steer clear of the chain restaurants...         2016   \n",
              "\n",
              "                                        clean_review  \\\n",
              "0  love go happy hour dinner great patio fan beat...   \n",
              "1  friend work connoisseur good food st louis are...   \n",
              "2  great service relax atmosphere food exceptiona...   \n",
              "3  I go brio first time wednesday dinner friend s...   \n",
              "4  I usually steer clear chain restaurant since m...   \n",
              "\n",
              "                                       clean_review2  food_similarity  \\\n",
              "0  [love, go, happy, hour, dinner, great, patio, ...         0.533379   \n",
              "1  [friend, work, connoisseur, good, food, st, lo...         0.603309   \n",
              "2  [great, service, relax, atmosphere, food, exce...         0.635701   \n",
              "3  [i, go, brio, first, time, wednesday, dinner, ...         0.614665   \n",
              "4  [i, usually, steer, clear, chain, restaurant, ...         0.565750   \n",
              "\n",
              "   service_similarity  place_similarity  price_similarity  \n",
              "0            0.550373          0.604214          0.501414  \n",
              "1            0.578850          0.592573          0.506964  \n",
              "2            0.723729          0.620446          0.501792  \n",
              "3            0.597369          0.633347          0.536175  \n",
              "4            0.524968          0.690019          0.501281  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-632c9717-40e3-451f-a5dc-981a58763246\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>stars</th>\n",
              "      <th>reviews</th>\n",
              "      <th>review_year</th>\n",
              "      <th>clean_review</th>\n",
              "      <th>clean_review2</th>\n",
              "      <th>food_similarity</th>\n",
              "      <th>service_similarity</th>\n",
              "      <th>place_similarity</th>\n",
              "      <th>price_similarity</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>2014</td>\n",
              "      <td>love go happy hour dinner great patio fan beat...</td>\n",
              "      <td>[love, go, happy, hour, dinner, great, patio, ...</td>\n",
              "      <td>0.533379</td>\n",
              "      <td>0.550373</td>\n",
              "      <td>0.604214</td>\n",
              "      <td>0.501414</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>5.0</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>2014</td>\n",
              "      <td>friend work connoisseur good food st louis are...</td>\n",
              "      <td>[friend, work, connoisseur, good, food, st, lo...</td>\n",
              "      <td>0.603309</td>\n",
              "      <td>0.578850</td>\n",
              "      <td>0.592573</td>\n",
              "      <td>0.506964</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>2015</td>\n",
              "      <td>great service relax atmosphere food exceptiona...</td>\n",
              "      <td>[great, service, relax, atmosphere, food, exce...</td>\n",
              "      <td>0.635701</td>\n",
              "      <td>0.723729</td>\n",
              "      <td>0.620446</td>\n",
              "      <td>0.501792</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>5.0</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>2015</td>\n",
              "      <td>I go brio first time wednesday dinner friend s...</td>\n",
              "      <td>[i, go, brio, first, time, wednesday, dinner, ...</td>\n",
              "      <td>0.614665</td>\n",
              "      <td>0.597369</td>\n",
              "      <td>0.633347</td>\n",
              "      <td>0.536175</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>4.0</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>2016</td>\n",
              "      <td>I usually steer clear chain restaurant since m...</td>\n",
              "      <td>[i, usually, steer, clear, chain, restaurant, ...</td>\n",
              "      <td>0.565750</td>\n",
              "      <td>0.524968</td>\n",
              "      <td>0.690019</td>\n",
              "      <td>0.501281</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-632c9717-40e3-451f-a5dc-981a58763246')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-632c9717-40e3-451f-a5dc-981a58763246 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-632c9717-40e3-451f-a5dc-981a58763246');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 86
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df['food_similarity']=df['food_similarity'].astype('float')"
      ],
      "metadata": {
        "id": "jCD0TOJbOO7b"
      },
      "execution_count": 87,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "#filter food similarity\n",
        "\n",
        "def cat_similarity(x):\n",
        "    if x > 0.60:\n",
        "        return '1'\n",
        "    else:\n",
        "        return '0'\n",
        "\n",
        "df['food'] = df['food_similarity'].apply(cat_similarity)\n",
        "\n",
        "df['food_similarity'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "OyDZzPDcGW9f",
        "outputId": "2f7ef3fc-2a38-4f56-b9f2-206f6f08ab7d"
      },
      "execution_count": 104,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.613320    5\n",
              "0.573721    4\n",
              "0.551213    4\n",
              "0.590407    4\n",
              "0.611379    4\n",
              "           ..\n",
              "0.540699    1\n",
              "0.567810    1\n",
              "0.563687    1\n",
              "0.648505    1\n",
              "0.518570    1\n",
              "Name: food_similarity, Length: 276426, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 104
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import seaborn as sns\n",
        "\n",
        "sns.countplot(x = \"food\", data = df)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "Ztqht5hYXv5Q",
        "outputId": "1c919234-f3f1-49e7-e8c6-930a32c7eea2"
      },
      "execution_count": 105,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAEGCAYAAABYV4NmAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAW80lEQVR4nO3df7BfdZ3f8efLRKzdXQpIStkkNKhZZyK1UTLIdFfHFYXAtAZd10K3El3G6ADtOnVa0c4Ui9LR7rrOYhUnSpZkxwWpiKRtKGZYV+rUKBdh+CnlErEkjSQGBLvsYoPv/vH9XPwS7r1cIJ/vF26ej5kz93zf5/M553NmMnnNOefzPd9UFZIkHWgvGvcAJEnzkwEjSerCgJEkdWHASJK6MGAkSV0sHPcAni+OPPLIWrZs2biHIUkvKDfddNNPqmrRdNsMmGbZsmVMTEyMexiS9IKS5EczbfMWmSSpCwNGktSFASNJ6qJbwCRZmuSbSe5MckeSP2j1I5JsTXJP+3t4qyfJxUkmk9ya5HVD+1rb2t+TZO1Q/fgkt7U+FyfJbMeQJI1OzyuYfcCHqmoFcCJwbpIVwPnA9VW1HLi+fQY4FVjelnXAJTAIC+AC4PXACcAFQ4FxCfC+oX6rW32mY0iSRqRbwFTVrqr6flv/GXAXsBhYA2xszTYCp7f1NcCmGtgGHJbkaOAUYGtVPVhVDwFbgdVt26FVta0Gb+zctN++pjuGJGlERvIMJsky4LXAd4GjqmpX2/Rj4Ki2vhi4f6jbjlabrb5jmjqzHGP/ca1LMpFkYs+ePc/8xCRJM+oeMEl+FbgK+GBVPTK8rV15dP29gNmOUVXrq2pVVa1atGja7wlJkp6lrgGT5MUMwuXLVfW1Vn6g3d6i/d3d6juBpUPdl7TabPUl09RnO4YkaUS6fZO/zei6FLirqv54aNNmYC3wyfb3mqH6eUmuYPBA/+Gq2pXkOuA/DD3YPxn4SFU9mOSRJCcyuPV2FvDZpzlGV8f/602jOIxeYG76w7PGPQRpLHq+KuY3gXcDtyW5pdU+yuA//SuTnA38CHhX27YFOA2YBB4F3gvQguTjwI2t3YVV9WBbPwe4DHgpcG1bmOUYkqQR6RYwVfVtIDNsPmma9gWcO8O+NgAbpqlPAMdNU9873TEkSaPjN/klSV0YMJKkLgwYSVIXBowkqQsDRpLUhQEjSerCgJEkdWHASJK6MGAkSV0YMJKkLgwYSVIXBowkqQsDRpLUhQEjSerCgJEkdWHASJK6MGAkSV10C5gkG5LsTnL7UO0rSW5py31TP6WcZFmSvx7a9oWhPscnuS3JZJKLk6TVj0iyNck97e/hrZ7WbjLJrUle1+scJUkz63kFcxmwerhQVf+0qlZW1UrgKuBrQ5vvndpWVR8Yql8CvA9Y3papfZ4PXF9Vy4Hr22eAU4farmv9JUkj1i1gquoG4MHptrWrkHcBl8+2jyRHA4dW1baqKmATcHrbvAbY2NY37lffVAPbgMPafiRJIzSuZzBvAB6oqnuGascmuTnJt5K8odUWAzuG2uxoNYCjqmpXW/8xcNRQn/tn6PMkSdYlmUgysWfPnudwOpKk/Y0rYM7kyVcvu4Bjquq1wL8C/jzJoXPdWbu6qWc6iKpaX1WrqmrVokWLnml3SdIsFo76gEkWAu8Ajp+qVdVjwGNt/aYk9wK/AewElgx1X9JqAA8kObqqdrVbYLtbfSewdIY+kqQRGccVzFuAH1TVE7e+kixKsqCtv5zBA/rt7RbYI0lObM9tzgKuad02A2vb+tr96me12WQnAg8P3UqTJI1Iz2nKlwPfAV6VZEeSs9umM3jqw/03Are2actfBT5QVVMTBM4BvgRMAvcC17b6J4G3JrmHQWh9stW3ANtb+y+2/pKkEet2i6yqzpyh/p5palcxmLY8XfsJ4Lhp6nuBk6apF3DuMxyuJOkA85v8kqQuDBhJUhcGjCSpCwNGktSFASNJ6sKAkSR1YcBIkrowYCRJXRgwkqQuDBhJUhcGjCSpCwNGktSFASNJ6sKAkSR1YcBIkrowYCRJXRgwkqQuev5k8oYku5PcPlT7WJKdSW5py2lD2z6SZDLJ3UlOGaqvbrXJJOcP1Y9N8t1W/0qSQ1r9Je3zZNu+rNc5SpJm1vMK5jJg9TT1z1TVyrZsAUiyAjgDeHXr8/kkC5IsAD4HnAqsAM5sbQE+1fb1SuAh4OxWPxt4qNU/09pJkkasW8BU1Q3Ag3Nsvga4oqoeq6ofApPACW2ZrKrtVfVz4ApgTZIAbwa+2vpvBE4f2tfGtv5V4KTWXpI0QuN4BnNeklvbLbTDW20xcP9Qmx2tNlP9ZcBPq2rffvUn7attf7i1f4ok65JMJJnYs2fPcz8zSdITRh0wlwCvAFYCu4BPj/j4T1JV66tqVVWtWrRo0TiHIknzzkgDpqoeqKrHq+oXwBcZ3AID2AksHWq6pNVmqu8FDkuycL/6k/bVtv+d1l6SNEIjDZgkRw99fDswNcNsM3BGmwF2LLAc+B5wI7C8zRg7hMFEgM1VVcA3gXe2/muBa4b2tbatvxP4i9ZekjRCC5++ybOT5HLgTcCRSXYAFwBvSrISKOA+4P0AVXVHkiuBO4F9wLlV9Xjbz3nAdcACYENV3dEO8WHgiiSfAG4GLm31S4E/SzLJYJLBGb3OUZI0s24BU1VnTlO+dJraVPuLgIumqW8BtkxT384vb7EN1/8G+N1nNFhJ0gHnN/klSV0YMJKkLgwYSVIXBowkqQsDRpLUhQEjSerCgJEkdWHASJK6MGAkSV0YMJKkLgwYSVIXBowkqQsDRpLUhQEjSerCgJEkdWHASJK6MGAkSV10C5gkG5LsTnL7UO0Pk/wgya1Jrk5yWKsvS/LXSW5pyxeG+hyf5LYkk0kuTpJWPyLJ1iT3tL+Ht3pau8l2nNf1OkdJ0sx6XsFcBqzer7YVOK6qXgP8L+AjQ9vuraqVbfnAUP0S4H3A8rZM7fN84PqqWg5c3z4DnDrUdl3rL0kasW4BU1U3AA/uV/tGVe1rH7cBS2bbR5KjgUOraltVFbAJOL1tXgNsbOsb96tvqoFtwGFtP5KkERrnM5jfB64d+nxskpuTfCvJG1ptMbBjqM2OVgM4qqp2tfUfA0cN9bl/hj6SpBFZOI6DJvm3wD7gy620CzimqvYmOR74epJXz3V/VVVJ6lmMYx2D22gcc8wxz7S7JGkWI7+CSfIe4B8Dv9due1FVj1XV3rZ+E3Av8BvATp58G21JqwE8MHXrq/3d3eo7gaUz9HmSqlpfVauqatWiRYsOwNlJkqaMNGCSrAb+DfC2qnp0qL4oyYK2/nIGD+i3t1tgjyQ5sc0eOwu4pnXbDKxt62v3q5/VZpOdCDw8dCtNkjQi3W6RJbkceBNwZJIdwAUMZo29BNjaZhtvazPG3ghcmOT/Ab8APlBVUxMEzmEwI+2lDJ7ZTD23+SRwZZKzgR8B72r1LcBpwCTwKPDeXucoSZpZt4CpqjOnKV86Q9urgKtm2DYBHDdNfS9w0jT1As59RoOVJB1wfpNfktSFASNJ6sKAkSR1YcBIkrowYCRJXRgwkqQu5hQwSa6fS02SpCmzfg8myd8C/jaDL0seDqRtOhRfIClJmsXTfdHy/cAHgV8HbuKXAfMI8J86jkuS9AI3a8BU1Z8Af5LkX1TVZ0c0JknSPDCnV8VU1WeT/CNg2XCfqtrUaVySpBe4OQVMkj8DXgHcAjzeylO/MClJ0lPM9WWXq4AVU7/fIknS05nr92BuB/5ez4FIkuaXuV7BHAncmeR7wGNTxap6W5dRSZJe8OYaMB/rOQhJ0vwz11lk3+o9EEnS/DLXWWQ/YzBrDOAQ4MXAX1XVob0GJkl6YZvTQ/6q+rWqOrQFykuB3wE+/3T9kmxIsjvJ7UO1I5JsTXJP+3t4qyfJxUkmk9ya5HVDfda29vckWTtUPz7Jba3PxUky2zEkSaPzjN+mXANfB06ZQ/PLgNX71c4Hrq+q5cD17TPAqcDytqwDLoFBWAAXAK8HTgAuGAqMS4D3DfVb/TTHkCSNyFxvkb1j6OOLGHwv5m+erl9V3ZBk2X7lNcCb2vpG4C+BD7f6pvZdm21JDktydGu7taoebGPZCqxO8pfAoVW1rdU3AacD185yDEnSiMx1Ftk/GVrfB9zH4D/xZ+OoqtrV1n8MHNXWFwP3D7Xb0Wqz1XdMU5/tGE+SZB2DqyWOOeaYZ3MukqQZzHUW2Xt7HLyqKknXtwPMdoyqWg+sB1i1apVvKZCkA2iuPzi2JMnV7YH97iRXJVnyLI/5QLv1Rfu7u9V3AkuH2i1ptdnqS6apz3YMSdKIzPUh/58Cmxn8LsyvA/+l1Z6NzcDUTLC1wDVD9bPabLITgYfbba7rgJOTHN4e7p8MXNe2PZLkxDZ77Kz99jXdMSRJIzLXgFlUVX9aVfvachmw6Ok6Jbkc+A7wqiQ7kpwNfBJ4a5J7gLe0zwBbgO3AJPBF4ByA9nD/48CNbblw6oF/a/Ol1udeBg/4meUYkqQRmetD/r1J/jlweft8JrD36TpV1ZkzbDppmrYFnDvDfjYAG6apTwDHTVPfO90xpIPV/77wH4x7CHoeOubf3dZ1/3O9gvl94F0MZmTtAt4JvKfTmCRJ88Bcr2AuBNZW1UPwxJcf/4hB8EiS9BRzvYJ5zVS4wBPPRV7bZ0iSpPlgrgHzouH3ebUrmLle/UiSDkJzDYlPA99J8p/b598FLuozJEnSfDDXb/JvSjIBvLmV3lFVd/YbliTphW7Ot7laoBgqkqQ5ecav65ckaS4MGElSFwaMJKkLA0aS1IUBI0nqwoCRJHVhwEiSujBgJEldGDCSpC4MGElSFyMPmCSvSnLL0PJIkg8m+ViSnUP104b6fCTJZJK7k5wyVF/dapNJzh+qH5vku63+lSSHjPo8JelgN/KAqaq7q2plVa0EjgceBa5umz8zta2qtgAkWQGcAbwaWA18PsmCJAuAzwGnAiuAM1tbgE+1fb0SeAg4e1TnJ0kaGPctspOAe6vqR7O0WQNcUVWPVdUPgUnghLZMVtX2qvo5cAWwJkkYvPX5q63/RuD0bmcgSZrWuAPmDODyoc/nJbk1yYahHzhbDNw/1GZHq81Ufxnw06rat1/9KZKsSzKRZGLPnj3P/WwkSU8YW8C05yJvA6Z+xOwS4BXASmAXgx8566qq1lfVqqpatWjRot6Hk6SDyjh/9vhU4PtV9QDA1F+AJF8E/mv7uBNYOtRvSasxQ30vcFiShe0qZri9JGlExnmL7EyGbo8lOXpo29uB29v6ZuCMJC9JciywHPgecCOwvM0YO4TB7bbNVVXAN4F3tv5rgWu6nokk6SnGcgWT5FeAtwLvHyr/xyQrgQLum9pWVXckuZLBr2nuA86tqsfbfs4DrgMWABuq6o62rw8DVyT5BHAzcGn3k5IkPclYAqaq/orBw/jh2rtnaX8RcNE09S3Almnq2xnMMpMkjcm4Z5FJkuYpA0aS1IUBI0nqwoCRJHVhwEiSujBgJEldGDCSpC4MGElSFwaMJKkLA0aS1IUBI0nqwoCRJHVhwEiSujBgJEldGDCSpC4MGElSFwaMJKmLsQVMkvuS3JbkliQTrXZEkq1J7ml/D2/1JLk4yWSSW5O8bmg/a1v7e5KsHaof3/Y/2fpm9GcpSQevcV/B/HZVrayqVe3z+cD1VbUcuL59BjgVWN6WdcAlMAgk4ALg9Qx+IvmCqVBqbd431G91/9ORJE0Zd8Dsbw2wsa1vBE4fqm+qgW3AYUmOBk4BtlbVg1X1ELAVWN22HVpV26qqgE1D+5IkjcA4A6aAbyS5Kcm6Vjuqqna19R8DR7X1xcD9Q313tNps9R3T1J8kybokE0km9uzZ81zPR5I0ZOEYj/1bVbUzyd8Ftib5wfDGqqok1XMAVbUeWA+watWqrseSpIPN2K5gqmpn+7sbuJrBM5QH2u0t2t/drflOYOlQ9yWtNlt9yTR1SdKIjCVgkvxKkl+bWgdOBm4HNgNTM8HWAte09c3AWW022YnAw+1W2nXAyUkObw/3Twaua9seSXJimz121tC+JEkjMK5bZEcBV7eZwwuBP6+q/57kRuDKJGcDPwLe1dpvAU4DJoFHgfcCVNWDST4O3NjaXVhVD7b1c4DLgJcC17ZFkjQiYwmYqtoO/MNp6nuBk6apF3DuDPvaAGyYpj4BHPecBytJelaeb9OUJUnzhAEjSerCgJEkdWHASJK6MGAkSV0YMJKkLgwYSVIXBowkqQsDRpLUhQEjSerCgJEkdWHASJK6MGAkSV0YMJKkLgwYSVIXBowkqQsDRpLUxcgDJsnSJN9McmeSO5L8Qat/LMnOJLe05bShPh9JMpnk7iSnDNVXt9pkkvOH6scm+W6rfyXJIaM9S0nSOK5g9gEfqqoVwInAuUlWtG2fqaqVbdkC0LadAbwaWA18PsmCJAuAzwGnAiuAM4f286m2r1cCDwFnj+rkJEkDIw+YqtpVVd9v6z8D7gIWz9JlDXBFVT1WVT8EJoET2jJZVdur6ufAFcCaJAHeDHy19d8InN7nbCRJMxnrM5gky4DXAt9tpfOS3JpkQ5LDW20xcP9Qtx2tNlP9ZcBPq2rffnVJ0giNLWCS/CpwFfDBqnoEuAR4BbAS2AV8egRjWJdkIsnEnj17eh9Okg4qYwmYJC9mEC5frqqvAVTVA1X1eFX9Avgig1tgADuBpUPdl7TaTPW9wGFJFu5Xf4qqWl9Vq6pq1aJFiw7MyUmSgPHMIgtwKXBXVf3xUP3ooWZvB25v65uBM5K8JMmxwHLge8CNwPI2Y+wQBhMBNldVAd8E3tn6rwWu6XlOkqSnWvj0TQ643wTeDdyW5JZW+yiDWWArgQLuA94PUFV3JLkSuJPBDLRzq+pxgCTnAdcBC4ANVXVH29+HgSuSfAK4mUGgSZJGaOQBU1XfBjLNpi2z9LkIuGia+pbp+lXVdn55i02SNAZ+k1+S1IUBI0nqwoCRJHVhwEiSujBgJEldGDCSpC4MGElSFwaMJKkLA0aS1IUBI0nqwoCRJHVhwEiSujBgJEldGDCSpC4MGElSFwaMJKkLA0aS1IUBI0nqYt4GTJLVSe5OMpnk/HGPR5IONvMyYJIsAD4HnAqsAM5MsmK8o5Kkg8u8DBjgBGCyqrZX1c+BK4A1Yx6TJB1UFo57AJ0sBu4f+rwDeP3+jZKsA9a1j/83yd0jGNvB4kjgJ+MexPNB/mjtuIegJ/Pf5pQLciD28vdn2jBfA2ZOqmo9sH7c45iPkkxU1apxj0Pan/82R2e+3iLbCSwd+ryk1SRJIzJfA+ZGYHmSY5McApwBbB7zmCTpoDIvb5FV1b4k5wHXAQuADVV1x5iHdbDx1qOer/y3OSKpqnGPQZI0D83XW2SSpDEzYCRJXRgwOqB8RY+er5JsSLI7ye3jHsvBwoDRAeMrevQ8dxmwetyDOJgYMDqQfEWPnreq6gbgwXGP42BiwOhAmu4VPYvHNBZJY2bASJK6MGB0IPmKHklPMGB0IPmKHklPMGB0wFTVPmDqFT13AVf6ih49XyS5HPgO8KokO5KcPe4xzXe+KkaS1IVXMJKkLgwYSVIXBowkqQsDRpLUhQEjSerCgJHGKMm/THJXki8/x/3cl+TIAzUu6UCYlz+ZLL2AnAO8pap2jHsg0oHmFYw0Jkm+ALwcuDbJh5J8PcmtSbYleU1rc8QM9Zcl+UaSO5J8CcgYT0WalgEjjUlVfQD4P8BvA8uAm6vqNcBHgU2t2b+foX4B8O2qejVwNXDMCIcuzYm3yKTnh98Cfgegqv6iXaEcOkv9jcA7Wv2/JXloTOOWZuQVjCSpCwNGen74H8DvASR5E/CTqnpklvoNwD9r9VOBw0c/ZGl2vuxSGqMk9wGrgF8AGxg89H8UWFdVtyY5Yob6y4DLGfxi6P8ETgaOr6qfjP4spOkZMJKkLrxFJknqwoCRJHVhwEiSujBgJEldGDCSpC4MGElSFwaMJKmL/w/R4kG2zuHFOAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#filter service similarity\n",
        "def cat_similarity(x):\n",
        "    if x > 0.60:\n",
        "        return '1'\n",
        "    else:\n",
        "        return '0'\n",
        "\n",
        "df['service'] = df['service_similarity'].apply(cat_similarity)\n",
        "\n",
        "df['service_similarity'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2G2FacDJH7oQ",
        "outputId": "ff05d5b8-bd47-43cb-a356-244b97b7d8ff"
      },
      "execution_count": 106,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.566939    4\n",
              "0.520032    4\n",
              "0.555218    4\n",
              "0.508155    4\n",
              "0.567525    4\n",
              "           ..\n",
              "0.455987    1\n",
              "0.549431    1\n",
              "0.608109    1\n",
              "0.522804    1\n",
              "0.446904    1\n",
              "Name: service_similarity, Length: 281965, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 106
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sns.countplot(x = \"service\", data = df)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "xxC9M3UVYcuo",
        "outputId": "cc76b452-0a44-4417-f7a5-5c74ad34dc06"
      },
      "execution_count": 107,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#filter place similarity\n",
        "def cat_similarity(x):\n",
        "    if x > 0.60:\n",
        "        return '1'\n",
        "    else:\n",
        "        return '0'\n",
        "\n",
        "df['place'] = df['place_similarity'].apply(cat_similarity)\n",
        "\n",
        "df['place_similarity'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EzPoeOR3INp_",
        "outputId": "39d379c4-79fc-439a-df78-2f27f2130c3d"
      },
      "execution_count": 108,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.590413    4\n",
              "0.586422    4\n",
              "0.567581    4\n",
              "0.596857    4\n",
              "0.609993    4\n",
              "           ..\n",
              "0.523070    1\n",
              "0.581824    1\n",
              "0.487596    1\n",
              "0.538697    1\n",
              "0.549674    1\n",
              "Name: place_similarity, Length: 277881, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 108
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sns.countplot(x = \"place\", data = df)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "ReAeN0umYpG2",
        "outputId": "c7405bfb-4a20-4f29-adb1-c7c7ad1a86a8"
      },
      "execution_count": 109,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#filter price similarity\n",
        "def cat_similarity(x):\n",
        "    if x > 0.60:\n",
        "        return '1'\n",
        "    else:\n",
        "        return '0'\n",
        "\n",
        "df['price'] = df['price_similarity'].apply(cat_similarity)\n",
        "\n",
        "df['price_similarity'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fDg4T5ewIVnV",
        "outputId": "65cbab2c-5fa2-4ba0-a4ba-cff283cd1366"
      },
      "execution_count": 110,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.514374    4\n",
              "0.504477    4\n",
              "0.519188    4\n",
              "0.520334    4\n",
              "0.532638    4\n",
              "           ..\n",
              "0.495542    1\n",
              "0.584874    1\n",
              "0.541759    1\n",
              "0.529147    1\n",
              "0.516854    1\n",
              "Name: price_similarity, Length: 279145, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 110
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "sns.countplot(x = \"price\", data = df)\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 279
        },
        "id": "82m55Pz-Zt3W",
        "outputId": "4975f86f-095f-413d-df01-1bb40c8e3627"
      },
      "execution_count": 111,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#add sentiment\n",
        "Cust = []\n",
        "for i in df['stars']:\n",
        "    if (i == 1):\n",
        "        Cust.append('BAD')\n",
        "    elif (i == 3) | (i == 2):\n",
        "        Cust.append('NEUTRAL')\n",
        "    else:\n",
        "        Cust.append('GOOD')\n",
        "        \n",
        "\n",
        "df['Customer EXP'] = Cust\n",
        "df['Customer EXP'].value_counts()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wa8C33g9FsJF",
        "outputId": "cfbf5860-e9cb-48f4-e258-be9d96e7aaab"
      },
      "execution_count": 112,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "GOOD       197895\n",
              "NEUTRAL     61608\n",
              "BAD         32873\n",
              "Name: Customer EXP, dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 112
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 583
        },
        "id": "TcKvjSyPHz27",
        "outputId": "067fafd0-f74b-42c0-90ab-5a07d2ff59d3"
      },
      "execution_count": 113,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  stars  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01    4.0   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26    5.0   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11    4.0   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25    5.0   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32    4.0   \n",
              "\n",
              "                                             reviews  review_year  \\\n",
              "0  Love going here for happy hour or dinner!  Gre...         2014   \n",
              "1  My friends at work (connoisseurs of good food ...         2014   \n",
              "2  Great service, relaxing atmosphere and the foo...         2015   \n",
              "3  I went to Brio for the first time on Wednesday...         2015   \n",
              "4  I usually steer clear of the chain restaurants...         2016   \n",
              "\n",
              "                                        clean_review  \\\n",
              "0  love go happy hour dinner great patio fan beat...   \n",
              "1  friend work connoisseur good food st louis are...   \n",
              "2  great service relax atmosphere food exceptiona...   \n",
              "3  I go brio first time wednesday dinner friend s...   \n",
              "4  I usually steer clear chain restaurant since m...   \n",
              "\n",
              "                                       clean_review2  food_similarity  \\\n",
              "0  [love, go, happy, hour, dinner, great, patio, ...         0.533379   \n",
              "1  [friend, work, connoisseur, good, food, st, lo...         0.603309   \n",
              "2  [great, service, relax, atmosphere, food, exce...         0.635701   \n",
              "3  [i, go, brio, first, time, wednesday, dinner, ...         0.614665   \n",
              "4  [i, usually, steer, clear, chain, restaurant, ...         0.565750   \n",
              "\n",
              "   service_similarity  place_similarity  price_similarity food service place  \\\n",
              "0            0.550373          0.604214          0.501414    0       0     1   \n",
              "1            0.578850          0.592573          0.506964    1       0     0   \n",
              "2            0.723729          0.620446          0.501792    1       1     1   \n",
              "3            0.597369          0.633347          0.536175    1       0     1   \n",
              "4            0.524968          0.690019          0.501281    0       0     1   \n",
              "\n",
              "  price Customer EXP  \n",
              "0     0         GOOD  \n",
              "1     0         GOOD  \n",
              "2     0         GOOD  \n",
              "3     0         GOOD  \n",
              "4     0         GOOD  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-da64d0ce-e5ab-4fd3-afed-0e7e020a99ad\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>stars</th>\n",
              "      <th>reviews</th>\n",
              "      <th>review_year</th>\n",
              "      <th>clean_review</th>\n",
              "      <th>clean_review2</th>\n",
              "      <th>food_similarity</th>\n",
              "      <th>service_similarity</th>\n",
              "      <th>place_similarity</th>\n",
              "      <th>price_similarity</th>\n",
              "      <th>food</th>\n",
              "      <th>service</th>\n",
              "      <th>place</th>\n",
              "      <th>price</th>\n",
              "      <th>Customer EXP</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>2014</td>\n",
              "      <td>love go happy hour dinner great patio fan beat...</td>\n",
              "      <td>[love, go, happy, hour, dinner, great, patio, ...</td>\n",
              "      <td>0.533379</td>\n",
              "      <td>0.550373</td>\n",
              "      <td>0.604214</td>\n",
              "      <td>0.501414</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>5.0</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>2014</td>\n",
              "      <td>friend work connoisseur good food st louis are...</td>\n",
              "      <td>[friend, work, connoisseur, good, food, st, lo...</td>\n",
              "      <td>0.603309</td>\n",
              "      <td>0.578850</td>\n",
              "      <td>0.592573</td>\n",
              "      <td>0.506964</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>2015</td>\n",
              "      <td>great service relax atmosphere food exceptiona...</td>\n",
              "      <td>[great, service, relax, atmosphere, food, exce...</td>\n",
              "      <td>0.635701</td>\n",
              "      <td>0.723729</td>\n",
              "      <td>0.620446</td>\n",
              "      <td>0.501792</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>5.0</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>2015</td>\n",
              "      <td>I go brio first time wednesday dinner friend s...</td>\n",
              "      <td>[i, go, brio, first, time, wednesday, dinner, ...</td>\n",
              "      <td>0.614665</td>\n",
              "      <td>0.597369</td>\n",
              "      <td>0.633347</td>\n",
              "      <td>0.536175</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>4.0</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>2016</td>\n",
              "      <td>I usually steer clear chain restaurant since m...</td>\n",
              "      <td>[i, usually, steer, clear, chain, restaurant, ...</td>\n",
              "      <td>0.565750</td>\n",
              "      <td>0.524968</td>\n",
              "      <td>0.690019</td>\n",
              "      <td>0.501281</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-da64d0ce-e5ab-4fd3-afed-0e7e020a99ad')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-da64d0ce-e5ab-4fd3-afed-0e7e020a99ad button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-da64d0ce-e5ab-4fd3-afed-0e7e020a99ad');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 113
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df.to_csv(\"/content/gdrive/MyDrive/yelp_dataset/Clean_Yelp_ABSA_Reviews2.csv\", index = False)"
      ],
      "metadata": {
        "id": "fdhWIWW1Anfg"
      },
      "execution_count": 114,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "# Model to predict sentiments on the basis of these aspects."
      ],
      "metadata": {
        "id": "dyQc4wpnLe2-"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import torch\n",
        "from torch import nn\n",
        "from torch.utils.data import Dataset\n",
        "from torch.utils.data import DataLoader\n",
        "from torch.utils.data import RandomSampler\n",
        "import warnings"
      ],
      "metadata": {
        "id": "0uillxL6AndO"
      },
      "execution_count": 117,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Configs"
      ],
      "metadata": {
        "id": "ZVRzbUbUMrWi"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class config:\n",
        "    warnings.filterwarnings(\"ignore\", category = UserWarning)\n",
        "    IMG_SIZE = (224,224)\n",
        "    DEVICE = (\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
        "    FOLDS = 5\n",
        "    SHUFFLE = True\n",
        "    BATCH_SIZE = 32\n",
        "    LR = 0.01\n",
        "    EPOCHS = 30\n",
        "    EMB_DIM = 100\n",
        "    MAX_LEN = 20\n",
        "    MODEL_PATH = \"./Models/MyModel.pt\"\n"
      ],
      "metadata": {
        "id": "ODbvZImNLWSb"
      },
      "execution_count": 118,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_absa = pd.read_csv(\"/content/gdrive/MyDrive/yelp_dataset/Clean_Yelp_ABSA_Reviews2.csv\")\n",
        "df_absa.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 583
        },
        "id": "6ZT6gtZoLWMM",
        "outputId": "34597906-9e2c-44fc-95ac-74430d8b6892"
      },
      "execution_count": 119,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  stars  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01    4.0   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26    5.0   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11    4.0   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25    5.0   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32    4.0   \n",
              "\n",
              "                                             reviews  review_year  \\\n",
              "0  Love going here for happy hour or dinner!  Gre...         2014   \n",
              "1  My friends at work (connoisseurs of good food ...         2014   \n",
              "2  Great service, relaxing atmosphere and the foo...         2015   \n",
              "3  I went to Brio for the first time on Wednesday...         2015   \n",
              "4  I usually steer clear of the chain restaurants...         2016   \n",
              "\n",
              "                                        clean_review  \\\n",
              "0  love go happy hour dinner great patio fan beat...   \n",
              "1  friend work connoisseur good food st louis are...   \n",
              "2  great service relax atmosphere food exceptiona...   \n",
              "3  I go brio first time wednesday dinner friend s...   \n",
              "4  I usually steer clear chain restaurant since m...   \n",
              "\n",
              "                                       clean_review2  food_similarity  \\\n",
              "0  ['love', 'go', 'happy', 'hour', 'dinner', 'gre...         0.533379   \n",
              "1  ['friend', 'work', 'connoisseur', 'good', 'foo...         0.603309   \n",
              "2  ['great', 'service', 'relax', 'atmosphere', 'f...         0.635701   \n",
              "3  ['i', 'go', 'brio', 'first', 'time', 'wednesda...         0.614665   \n",
              "4  ['i', 'usually', 'steer', 'clear', 'chain', 'r...         0.565750   \n",
              "\n",
              "   service_similarity  place_similarity  price_similarity  food  service  \\\n",
              "0            0.550373          0.604214          0.501414     0        0   \n",
              "1            0.578850          0.592573          0.506964     1        0   \n",
              "2            0.723729          0.620446          0.501792     1        1   \n",
              "3            0.597369          0.633347          0.536175     1        0   \n",
              "4            0.524968          0.690019          0.501281     0        0   \n",
              "\n",
              "   place  price Customer EXP  \n",
              "0      1      0         GOOD  \n",
              "1      0      0         GOOD  \n",
              "2      1      0         GOOD  \n",
              "3      1      0         GOOD  \n",
              "4      1      0         GOOD  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-b73c01df-1668-44b4-95ab-2ab958b1a8b0\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>stars</th>\n",
              "      <th>reviews</th>\n",
              "      <th>review_year</th>\n",
              "      <th>clean_review</th>\n",
              "      <th>clean_review2</th>\n",
              "      <th>food_similarity</th>\n",
              "      <th>service_similarity</th>\n",
              "      <th>place_similarity</th>\n",
              "      <th>price_similarity</th>\n",
              "      <th>food</th>\n",
              "      <th>service</th>\n",
              "      <th>place</th>\n",
              "      <th>price</th>\n",
              "      <th>Customer EXP</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>2014</td>\n",
              "      <td>love go happy hour dinner great patio fan beat...</td>\n",
              "      <td>['love', 'go', 'happy', 'hour', 'dinner', 'gre...</td>\n",
              "      <td>0.533379</td>\n",
              "      <td>0.550373</td>\n",
              "      <td>0.604214</td>\n",
              "      <td>0.501414</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>5.0</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>2014</td>\n",
              "      <td>friend work connoisseur good food st louis are...</td>\n",
              "      <td>['friend', 'work', 'connoisseur', 'good', 'foo...</td>\n",
              "      <td>0.603309</td>\n",
              "      <td>0.578850</td>\n",
              "      <td>0.592573</td>\n",
              "      <td>0.506964</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>2015</td>\n",
              "      <td>great service relax atmosphere food exceptiona...</td>\n",
              "      <td>['great', 'service', 'relax', 'atmosphere', 'f...</td>\n",
              "      <td>0.635701</td>\n",
              "      <td>0.723729</td>\n",
              "      <td>0.620446</td>\n",
              "      <td>0.501792</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>5.0</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>2015</td>\n",
              "      <td>I go brio first time wednesday dinner friend s...</td>\n",
              "      <td>['i', 'go', 'brio', 'first', 'time', 'wednesda...</td>\n",
              "      <td>0.614665</td>\n",
              "      <td>0.597369</td>\n",
              "      <td>0.633347</td>\n",
              "      <td>0.536175</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>4.0</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>2016</td>\n",
              "      <td>I usually steer clear chain restaurant since m...</td>\n",
              "      <td>['i', 'usually', 'steer', 'clear', 'chain', 'r...</td>\n",
              "      <td>0.565750</td>\n",
              "      <td>0.524968</td>\n",
              "      <td>0.690019</td>\n",
              "      <td>0.501281</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-b73c01df-1668-44b4-95ab-2ab958b1a8b0')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-b73c01df-1668-44b4-95ab-2ab958b1a8b0 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-b73c01df-1668-44b4-95ab-2ab958b1a8b0');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 119
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_absa.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 364
        },
        "id": "SLOVdmFIK85W",
        "outputId": "e61c1d38-5c86-4c05-f2ea-7ec436d60b38"
      },
      "execution_count": 120,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "               stars    review_year  food_similarity  service_similarity  \\\n",
              "count  292376.000000  292376.000000    292376.000000       292376.000000   \n",
              "mean        3.787113    2016.096964         0.577500            0.544896   \n",
              "std         1.368116       3.196996         0.047507            0.067984   \n",
              "min         1.000000    2005.000000         0.000000           -0.046627   \n",
              "25%         3.000000    2014.000000         0.549304            0.500160   \n",
              "50%         4.000000    2016.000000         0.580594            0.542505   \n",
              "75%         5.000000    2019.000000         0.609243            0.587919   \n",
              "max         5.000000    2022.000000         0.768563            0.846891   \n",
              "\n",
              "       place_similarity  price_similarity           food        service  \\\n",
              "count     292376.000000     292376.000000  292376.000000  292376.000000   \n",
              "mean           0.596461          0.508878       0.325222       0.199356   \n",
              "std            0.051792          0.048489       0.468458       0.399517   \n",
              "min            0.000000          0.000000       0.000000       0.000000   \n",
              "25%            0.564553          0.480152       0.000000       0.000000   \n",
              "50%            0.599314          0.508434       0.000000       0.000000   \n",
              "75%            0.631412          0.536732       1.000000       0.000000   \n",
              "max            0.817265          0.778234       1.000000       1.000000   \n",
              "\n",
              "               place          price  \n",
              "count  292376.000000  292376.000000  \n",
              "mean        0.494363       0.035492  \n",
              "std         0.499969       0.185020  \n",
              "min         0.000000       0.000000  \n",
              "25%         0.000000       0.000000  \n",
              "50%         0.000000       0.000000  \n",
              "75%         1.000000       0.000000  \n",
              "max         1.000000       1.000000  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-233f9361-964a-405b-8f27-c0e93e9f79c2\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>stars</th>\n",
              "      <th>review_year</th>\n",
              "      <th>food_similarity</th>\n",
              "      <th>service_similarity</th>\n",
              "      <th>place_similarity</th>\n",
              "      <th>price_similarity</th>\n",
              "      <th>food</th>\n",
              "      <th>service</th>\n",
              "      <th>place</th>\n",
              "      <th>price</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "      <td>292376.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>3.787113</td>\n",
              "      <td>2016.096964</td>\n",
              "      <td>0.577500</td>\n",
              "      <td>0.544896</td>\n",
              "      <td>0.596461</td>\n",
              "      <td>0.508878</td>\n",
              "      <td>0.325222</td>\n",
              "      <td>0.199356</td>\n",
              "      <td>0.494363</td>\n",
              "      <td>0.035492</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>1.368116</td>\n",
              "      <td>3.196996</td>\n",
              "      <td>0.047507</td>\n",
              "      <td>0.067984</td>\n",
              "      <td>0.051792</td>\n",
              "      <td>0.048489</td>\n",
              "      <td>0.468458</td>\n",
              "      <td>0.399517</td>\n",
              "      <td>0.499969</td>\n",
              "      <td>0.185020</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>1.000000</td>\n",
              "      <td>2005.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>-0.046627</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>3.000000</td>\n",
              "      <td>2014.000000</td>\n",
              "      <td>0.549304</td>\n",
              "      <td>0.500160</td>\n",
              "      <td>0.564553</td>\n",
              "      <td>0.480152</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>4.000000</td>\n",
              "      <td>2016.000000</td>\n",
              "      <td>0.580594</td>\n",
              "      <td>0.542505</td>\n",
              "      <td>0.599314</td>\n",
              "      <td>0.508434</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>5.000000</td>\n",
              "      <td>2019.000000</td>\n",
              "      <td>0.609243</td>\n",
              "      <td>0.587919</td>\n",
              "      <td>0.631412</td>\n",
              "      <td>0.536732</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>5.000000</td>\n",
              "      <td>2022.000000</td>\n",
              "      <td>0.768563</td>\n",
              "      <td>0.846891</td>\n",
              "      <td>0.817265</td>\n",
              "      <td>0.778234</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-233f9361-964a-405b-8f27-c0e93e9f79c2')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-233f9361-964a-405b-8f27-c0e93e9f79c2 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-233f9361-964a-405b-8f27-c0e93e9f79c2');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 120
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Dataset Generator"
      ],
      "metadata": {
        "id": "hjPJqfYmMuWV"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "class Vocabulary:\n",
        "  \n",
        "    '''\n",
        "    __init__ method is called by default as soon as an object of this class is initiated\n",
        "    we use this method to initiate our vocab dictionaries\n",
        "    '''\n",
        "    def __init__(self, freq_threshold, max_size):\n",
        "        '''\n",
        "        freq_threshold : the minimum times a word must occur in corpus to be treated in vocab\n",
        "        max_size : max source vocab size. Eg. if set to 10,000, we pick the top 10,000 most frequent words and discard others\n",
        "        '''\n",
        "        #initiate the index to token dict\n",
        "        ## PAD -> padding, used for padding the shorter sentences in a batch to match the length of longest sentence in the batch\n",
        "        ## SOS -> start token, added in front of each sentence to signify the start of sentence\n",
        "        ## EOS -> End of sentence token, added to the end of each sentence to signify the end of sentence\n",
        "        ## UNK -> words which are not found in the vocab are replace by this token\n",
        "        self.itos = {0: 'padding', 1:'start', 2:'end', 3: 'unknown'}\n",
        "        #initiate the token to index dict\n",
        "        self.stoi = {k:j for j,k in self.itos.items()} \n",
        "        \n",
        "        self.freq_threshold = freq_threshold\n",
        "        self.max_size = max_size\n",
        "    \n",
        "    '''\n",
        "    __len__ is used by dataloader later to create batches\n",
        "    '''\n",
        "    def __len__(self):\n",
        "        return len(self.itos)\n",
        "    \n",
        "    '''\n",
        "    a simple tokenizer to split on space and converts the sentence to list of words\n",
        "    '''\n",
        "    @staticmethod\n",
        "    def tokenizer(text):\n",
        "        return [tok.lower().strip() for tok in text.split(' ')]\n",
        "    \n",
        "    '''\n",
        "    build the vocab: create a dictionary mapping of index to string (itos) and string to index (stoi)\n",
        "    output ex. for stoi -> {'the':5, 'a':6, 'an':7}\n",
        "    '''\n",
        "    def build_vocabulary(self, sentence_list):\n",
        "        #calculate the frequencies of each word first to remove the words with freq < freq_threshold\n",
        "        frequencies = {}  #init the freq dict\n",
        "        idx = 4 #index from which we want our dict to start. We already used 4 indexes for pad, start, end, unk\n",
        "        \n",
        "        #calculate freq of words\n",
        "        for sentence in sentence_list:\n",
        "            for word in self.tokenizer(sentence):\n",
        "                if word not in frequencies.keys():\n",
        "                    frequencies[word]=1\n",
        "                else:\n",
        "                    frequencies[word]+=1\n",
        "                    \n",
        "                    \n",
        "        #limit vocab by removing low freq words\n",
        "        frequencies = {k:v for k,v in frequencies.items() if v>self.freq_threshold} \n",
        "        \n",
        "        #limit vocab to the max_size specified\n",
        "        frequencies = dict(sorted(frequencies.items(), key = lambda x: -x[1])[:self.max_size-idx]) # idx =4 for pad, start, end , unk\n",
        "            \n",
        "        #create vocab\n",
        "        for word in frequencies.keys():\n",
        "            self.stoi[word] = idx\n",
        "            self.itos[idx] = word\n",
        "            idx+=1\n",
        "            \n",
        "     \n",
        "    '''\n",
        "    convert the list of words to a list of corresponding indexes\n",
        "    '''    \n",
        "    def numericalize(self, text):\n",
        "        #tokenize text\n",
        "        tokenized_text = self.tokenizer(text)\n",
        "        numericalized_text = []\n",
        "        for token in tokenized_text:\n",
        "            if token in self.stoi.keys():\n",
        "                numericalized_text.append(self.stoi[token])\n",
        "            else: #out-of-vocab words are represented by UNK token index\n",
        "                numericalized_text.append(self.stoi['unknown'])\n",
        "                \n",
        "        return numericalized_text"
      ],
      "metadata": {
        "id": "_FHQ8W-PLWHW"
      },
      "execution_count": 136,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from torch.utils.data import Dataset\n",
        "\n",
        "class CustomDataset(Dataset):\n",
        "    '''\n",
        "    Initiating Variables\n",
        "    df: the training dataframe\n",
        "    source_column : the name of source text column in the dataframe\n",
        "    transform : If we want to add any augmentation\n",
        "    freq_threshold : the minimum times a word must occur in corpus to be treated in vocab\n",
        "    source_vocab_max_size : max source vocab size\n",
        "    '''\n",
        "    \n",
        "    def __init__(self, df, source_column,freq_threshold = 3,\n",
        "                source_vocab_max_size = 10000 , transform=None):\n",
        "    \n",
        "        self.df = df\n",
        "        self.transform = transform\n",
        "        \n",
        "        #get source and target texts\n",
        "        self.source_texts = self.df[source_column]\n",
        "        \n",
        "        \n",
        "        ##VOCAB class has been created above\n",
        "        #Initialize source vocab object and build vocabulary\n",
        "        self.source_vocab = Vocabulary(freq_threshold, source_vocab_max_size)\n",
        "        self.source_vocab.build_vocabulary(self.source_texts.tolist())\n",
        "\n",
        "        \n",
        "    def __len__(self):\n",
        "        return len(self.df)\n",
        "    \n",
        "    '''\n",
        "    __getitem__ runs on 1 example at a time. Here, we get an example at index and return its numericalize source and\n",
        "    target values using the vocabulary objects we created in __init__\n",
        "    '''\n",
        "    def __getitem__(self, index):\n",
        "        source_text = self.source_texts[index]\n",
        "        \n",
        "        if self.transform is not None:\n",
        "            source_text = self.transform(source_text)\n",
        "            \n",
        "        #numericalize texts ['<SOS>','cat', 'in', 'a', 'bag','<EOS>'] -> [1,12,2,9,24,2]\n",
        "        numerialized_source = [self.source_vocab.stoi[\"start\"]]\n",
        "        numerialized_source += self.source_vocab.numericalize(source_text)\n",
        "        numerialized_source.append(self.source_vocab.stoi[\"end\"])\n",
        "        \n",
        "        #convert the list to tensor and return\n",
        "        return torch.tensor(numerialized_source), torch.tensor(self.df.y[index])"
      ],
      "metadata": {
        "id": "8Yba_M4hLWFA"
      },
      "execution_count": 137,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_absa2 = df"
      ],
      "metadata": {
        "id": "4_qrcZ7DcL8I"
      },
      "execution_count": 141,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df_absa2.head()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 583
        },
        "id": "qn_h3iusLWCI",
        "outputId": "b21815e7-4f76-4dd9-c607-df9ff9d1cbbe"
      },
      "execution_count": 142,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  user_id             business_id                 date  stars  \\\n",
              "0  9OAtfnWag-ajVxRbUTGIyg  lj-E32x9_FA7GmUrBGBEWg  2014-06-27 22:44:01    4.0   \n",
              "1  5h9JA231vPilNAIjHxwGng  lj-E32x9_FA7GmUrBGBEWg  2014-08-24 19:24:26    5.0   \n",
              "2  c1fSI6Dv5lybr0AJh67e7w  lj-E32x9_FA7GmUrBGBEWg  2015-09-24 15:01:11    4.0   \n",
              "3  st-q1iyW3sJm-v0OCrheoQ  lj-E32x9_FA7GmUrBGBEWg  2015-10-17 04:56:25    5.0   \n",
              "4  9qBdzBzoDxLFSMhhGrTWJg  lj-E32x9_FA7GmUrBGBEWg  2016-01-04 16:56:32    4.0   \n",
              "\n",
              "                                             reviews  review_year  \\\n",
              "0  Love going here for happy hour or dinner!  Gre...         2014   \n",
              "1  My friends at work (connoisseurs of good food ...         2014   \n",
              "2  Great service, relaxing atmosphere and the foo...         2015   \n",
              "3  I went to Brio for the first time on Wednesday...         2015   \n",
              "4  I usually steer clear of the chain restaurants...         2016   \n",
              "\n",
              "                                        clean_review  \\\n",
              "0  love go happy hour dinner great patio fan beat...   \n",
              "1  friend work connoisseur good food st louis are...   \n",
              "2  great service relax atmosphere food exceptiona...   \n",
              "3  I go brio first time wednesday dinner friend s...   \n",
              "4  I usually steer clear chain restaurant since m...   \n",
              "\n",
              "                                       clean_review2  food_similarity  \\\n",
              "0  [love, go, happy, hour, dinner, great, patio, ...         0.533379   \n",
              "1  [friend, work, connoisseur, good, food, st, lo...         0.603309   \n",
              "2  [great, service, relax, atmosphere, food, exce...         0.635701   \n",
              "3  [i, go, brio, first, time, wednesday, dinner, ...         0.614665   \n",
              "4  [i, usually, steer, clear, chain, restaurant, ...         0.565750   \n",
              "\n",
              "   service_similarity  place_similarity  price_similarity food service place  \\\n",
              "0            0.550373          0.604214          0.501414    0       0     1   \n",
              "1            0.578850          0.592573          0.506964    1       0     0   \n",
              "2            0.723729          0.620446          0.501792    1       1     1   \n",
              "3            0.597369          0.633347          0.536175    1       0     1   \n",
              "4            0.524968          0.690019          0.501281    0       0     1   \n",
              "\n",
              "  price Customer EXP  \n",
              "0     0         GOOD  \n",
              "1     0         GOOD  \n",
              "2     0         GOOD  \n",
              "3     0         GOOD  \n",
              "4     0         GOOD  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-0adaea72-31fc-4d0b-9ecb-8dbb13aa363d\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>user_id</th>\n",
              "      <th>business_id</th>\n",
              "      <th>date</th>\n",
              "      <th>stars</th>\n",
              "      <th>reviews</th>\n",
              "      <th>review_year</th>\n",
              "      <th>clean_review</th>\n",
              "      <th>clean_review2</th>\n",
              "      <th>food_similarity</th>\n",
              "      <th>service_similarity</th>\n",
              "      <th>place_similarity</th>\n",
              "      <th>price_similarity</th>\n",
              "      <th>food</th>\n",
              "      <th>service</th>\n",
              "      <th>place</th>\n",
              "      <th>price</th>\n",
              "      <th>Customer EXP</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>9OAtfnWag-ajVxRbUTGIyg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-06-27 22:44:01</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Love going here for happy hour or dinner!  Gre...</td>\n",
              "      <td>2014</td>\n",
              "      <td>love go happy hour dinner great patio fan beat...</td>\n",
              "      <td>[love, go, happy, hour, dinner, great, patio, ...</td>\n",
              "      <td>0.533379</td>\n",
              "      <td>0.550373</td>\n",
              "      <td>0.604214</td>\n",
              "      <td>0.501414</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>5h9JA231vPilNAIjHxwGng</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2014-08-24 19:24:26</td>\n",
              "      <td>5.0</td>\n",
              "      <td>My friends at work (connoisseurs of good food ...</td>\n",
              "      <td>2014</td>\n",
              "      <td>friend work connoisseur good food st louis are...</td>\n",
              "      <td>[friend, work, connoisseur, good, food, st, lo...</td>\n",
              "      <td>0.603309</td>\n",
              "      <td>0.578850</td>\n",
              "      <td>0.592573</td>\n",
              "      <td>0.506964</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>c1fSI6Dv5lybr0AJh67e7w</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-09-24 15:01:11</td>\n",
              "      <td>4.0</td>\n",
              "      <td>Great service, relaxing atmosphere and the foo...</td>\n",
              "      <td>2015</td>\n",
              "      <td>great service relax atmosphere food exceptiona...</td>\n",
              "      <td>[great, service, relax, atmosphere, food, exce...</td>\n",
              "      <td>0.635701</td>\n",
              "      <td>0.723729</td>\n",
              "      <td>0.620446</td>\n",
              "      <td>0.501792</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>st-q1iyW3sJm-v0OCrheoQ</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2015-10-17 04:56:25</td>\n",
              "      <td>5.0</td>\n",
              "      <td>I went to Brio for the first time on Wednesday...</td>\n",
              "      <td>2015</td>\n",
              "      <td>I go brio first time wednesday dinner friend s...</td>\n",
              "      <td>[i, go, brio, first, time, wednesday, dinner, ...</td>\n",
              "      <td>0.614665</td>\n",
              "      <td>0.597369</td>\n",
              "      <td>0.633347</td>\n",
              "      <td>0.536175</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>9qBdzBzoDxLFSMhhGrTWJg</td>\n",
              "      <td>lj-E32x9_FA7GmUrBGBEWg</td>\n",
              "      <td>2016-01-04 16:56:32</td>\n",
              "      <td>4.0</td>\n",
              "      <td>I usually steer clear of the chain restaurants...</td>\n",
              "      <td>2016</td>\n",
              "      <td>I usually steer clear chain restaurant since m...</td>\n",
              "      <td>[i, usually, steer, clear, chain, restaurant, ...</td>\n",
              "      <td>0.565750</td>\n",
              "      <td>0.524968</td>\n",
              "      <td>0.690019</td>\n",
              "      <td>0.501281</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>GOOD</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0adaea72-31fc-4d0b-9ecb-8dbb13aa363d')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-0adaea72-31fc-4d0b-9ecb-8dbb13aa363d button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-0adaea72-31fc-4d0b-9ecb-8dbb13aa363d');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 142
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_absa2=df_absa.fillna(\"NONE\")"
      ],
      "metadata": {
        "id": "22---ApObJ0i"
      },
      "execution_count": 144,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Assigning labels to each review\n",
        "\n",
        "def label(stars):\n",
        "    if stars == 5:\n",
        "        return 1\n",
        "    elif stars == 4:\n",
        "        return 1\n",
        "    else:\n",
        "        return 0\n",
        "    \n",
        "from tqdm import tqdm\n",
        "\n",
        "tqdm.pandas()\n",
        "\n",
        "df_absa2['y'] = df_absa2.stars.progress_map(label)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UELrR9ipc8MN",
        "outputId": "3bd21cb8-8500-4771-9aa0-77d06a16f66f"
      },
      "execution_count": 155,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "100%|██████████| 292376/292376 [00:00<00:00, 674557.13it/s]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "dataset = CustomDataset(df_absa2, \"clean_review\")"
      ],
      "metadata": {
        "id": "mjlTqi8JLV44"
      },
      "execution_count": 156,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "len(dataset.source_vocab.stoi)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "yGz6xPSbL18m",
        "outputId": "0a9b4103-7e9f-4c43-f795-68661257cc8a"
      },
      "execution_count": 157,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "9997"
            ]
          },
          "metadata": {},
          "execution_count": 157
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import pickle\n",
        "\n",
        "with open('dataset-new', 'wb') as dataset_file:\n",
        " \n",
        "  # Step 3\n",
        "    pickle.dump(dataset, dataset_file, pickle.HIGHEST_PROTOCOL)\n",
        "\n",
        "# import pickle\n",
        " \n",
        "# # Step 2\n",
        "# with open('./dataset', 'rb') as config_dictionary_file:\n",
        " \n",
        "#     # Step 3\n",
        "#     config_dictionary = pickle.load(config_dictionary_file)\n",
        " \n",
        "#     # After config_dictionary is read from file\n",
        "#     print(config_dictionary)"
      ],
      "metadata": {
        "id": "-CCl2KocL1yD"
      },
      "execution_count": 158,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Word embeddings"
      ],
      "metadata": {
        "id": "Kq9zjPYOM0Nh"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def get_emb_layer_with_weights(target_vocab, emb_model, trainable = False):\n",
        "\n",
        "    weights_matrix = np.zeros((len(target_vocab), config.EMB_DIM))\n",
        "    words_found = 0\n",
        "    \n",
        "    for i, word in enumerate(target_vocab):\n",
        "        weights_matrix[i] = np.concatenate([emb_model.wv[word]])\n",
        "        words_found += 1\n",
        "                \n",
        "    print(f\"Words found are : {words_found}\")\n",
        "    \n",
        "    weights_matrix = torch.tensor(weights_matrix, dtype = torch.float32).reshape(len(target_vocab), config.EMB_DIM)\n",
        "    emb_layer = nn.Embedding.from_pretrained(weights_matrix)\n",
        "    print(emb_layer)\n",
        "    if trainable:\n",
        "        emb_layer.weight.requires_grad = True\n",
        "    else:\n",
        "        emb_layer.weight.requires_grad = False\n",
        "\n",
        "    return emb_layer"
      ],
      "metadata": {
        "id": "cK1QwRTOL1wi"
      },
      "execution_count": 159,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "class MyCollate:\n",
        "    def __init__(self, pad_idx, maxlen):\n",
        "        self.pad_idx = pad_idx\n",
        "        self.maxlen = maxlen\n",
        "        \n",
        "    \n",
        "    #__call__: a default method\n",
        "    ##   First the obj is created using MyCollate(pad_idx) in data loader\n",
        "    ##   Then if obj(batch) is called -> __call__ runs by default\n",
        "    def __call__(self, batch):\n",
        "        #get all source indexed sentences of the batch\n",
        "        source = [item[0] for item in batch] \n",
        "        #pad them using pad_sequence method from pytorch. \n",
        "#         source = pad_sequence(source, batch_first=False, padding_value = self.pad_idx)\n",
        "        \n",
        "        padded_sequence = torch.zeros((self.maxlen, len(batch)), dtype = torch.int)\n",
        "        \n",
        "        for idx, text in enumerate(source):\n",
        "            \n",
        "            if len(text) > self.maxlen:\n",
        "                padded_sequence[:, idx] = source[idx][: self.maxlen]\n",
        "            else:\n",
        "                padded_sequence[:len(source[idx]), idx] = padded_sequence[:len(source[idx]), idx] + source[idx]\n",
        "                \n",
        "        \n",
        "        #get all target indexed sentences of the batch\n",
        "        target = [item[1] for item in batch] \n",
        "        \n",
        "        target = torch.tensor(target, dtype = torch.float32).reshape(-1)\n",
        "        return padded_sequence, target"
      ],
      "metadata": {
        "id": "G1FQJbkcL1vC"
      },
      "execution_count": 160,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "class Model(nn.Module):\n",
        "    def __init__(self, input_dim, embedding_dim, hidden_dim, output_dim, embedding_layer):\n",
        "        super().__init__()\n",
        "#         self.embedding = nn.Embedding(input_dim, embedding_dim)\n",
        "        self.hidden_dim = hidden_dim\n",
        "        self.embedding = embedding_layer\n",
        "        self.lstm = nn.LSTM(embedding_dim, hidden_dim, bidirectional = True)\n",
        "        self.fc1 = nn.Linear(2*hidden_dim, 128)\n",
        "        self.fc2 = nn.Linear(128, output_dim)\n",
        "        self.dropout = nn.Dropout(0.3)\n",
        "        self.sigmoid = nn.Sigmoid()\n",
        "\n",
        "\n",
        "        \n",
        "    def forward(self, text):\n",
        "        \n",
        "        max_len, N = text.shape\n",
        "        hidden = torch.zeros((2, N , self.hidden_dim),\n",
        "                          dtype=torch.float)\n",
        "        memory = torch.zeros((2, N , self.hidden_dim),\n",
        "                          dtype=torch.float)\n",
        "        hidden = hidden.to(config.DEVICE)\n",
        "        memory = memory.to(config.DEVICE)\n",
        "        embedded = self.embedding(text)\n",
        "        output, hidden = self.lstm(embedded, (hidden, memory))\n",
        "#         assert torch.equal(output[-1,:,:], hidden.squeeze(0))\n",
        "        y_pred = output[-1,:,:]\n",
        "        y_pred = self.fc1(y_pred)\n",
        "        y_pred = self.fc2(y_pred)\n",
        "        y_pred = self.sigmoid(y_pred)\n",
        "                         \n",
        "        return y_pred  "
      ],
      "metadata": {
        "id": "nSqF1zdbL1qz"
      },
      "execution_count": 161,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Training and Kfold Cross Validation"
      ],
      "metadata": {
        "id": "Eb49BJTNNA9d"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def train_epochs(dataloader,model, loss_fn, optimizer):\n",
        "    train_correct = 0\n",
        "    train_loss = 0\n",
        "    \n",
        "    model.train()\n",
        "\n",
        "    for review, label in tqdm(dataloader):\n",
        "        \n",
        "        review, label = review.to(config.DEVICE), label.to(config.DEVICE)\n",
        "        optimizer.zero_grad()\n",
        "        output = model(review)\n",
        "        output = output.reshape(-1)\n",
        "        loss = loss_fn(output, label)\n",
        "        loss.backward()\n",
        "        optimizer.step()\n",
        "        \n",
        "        train_loss += loss.item()*review.size(1)\n",
        "        prediction = (output > 0.5).float()\n",
        "        train_correct += (prediction == label).float().sum()\n",
        "        \n",
        "    return train_loss, train_correct\n"
      ],
      "metadata": {
        "id": "y6wR-i-2L1j2"
      },
      "execution_count": 162,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def val_epochs(dataloader, model, loss_fn):\n",
        "    val_correct = 0\n",
        "    val_loss = 0\n",
        "    \n",
        "    model.eval()\n",
        "#     hidden = model.init_hidden(config.BATCH_SIZE)\n",
        "\n",
        "    for review, label in dataloader:\n",
        "        \n",
        "        review, label = review.to(config.DEVICE), label.to(config.DEVICE)\n",
        "        \n",
        "        output = model(review)\n",
        "        output = output.reshape(-1)\n",
        "\n",
        "        loss = loss_fn(output, label)\n",
        "        \n",
        "        val_loss += loss.item()*review.size(1)\n",
        "        prediction = (output > 0.5).float()\n",
        "        val_correct += (prediction == label).float().sum()\n",
        "#         prediction = \n",
        "    return val_loss, val_correct"
      ],
      "metadata": {
        "id": "T9UdOst3L1dK"
      },
      "execution_count": 163,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from sklearn.model_selection import StratifiedKFold\n",
        "from sklearn.model_selection import KFold\n",
        "from torch.utils.data import SubsetRandomSampler\n",
        "from torch.optim import Adam\n",
        "from tqdm import tqdm\n",
        "from torch.utils.data import DataLoader \n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "\n",
        "# sfk = StratifiedKFold(n_splits = config.FOLDS)\n",
        "kfold = KFold(n_splits = config.FOLDS)\n",
        "model_state_dicts = {}\n",
        " \n",
        "for fold, (train_idx, val_idx) in enumerate(kfold.split(np.arange(len(dataset)))):\n",
        "    \n",
        "    train_sampler = SubsetRandomSampler(train_idx)\n",
        "    val_sampler = SubsetRandomSampler(val_idx)\n",
        "    \n",
        "    train_loader = DataLoader(dataset, batch_size = config.BATCH_SIZE, sampler = train_sampler, collate_fn = MyCollate(0, config.MAX_LEN))\n",
        "    val_loader = DataLoader(dataset, batch_size = config.BATCH_SIZE, sampler = val_sampler, collate_fn = MyCollate(0, config.MAX_LEN))  \n",
        "    \n",
        "    VOCAB_SIZE = len(dataset.source_vocab)\n",
        "    HIDDEN_DIM = 128\n",
        "    OUTPUT_DIM = 1\n",
        "    VOCAB = list(dataset.source_vocab.stoi)\n",
        "\n",
        "    embedding_layer = get_emb_layer_with_weights(target_vocab = VOCAB, emb_model = fasttext_model, trainable = False)\n",
        "\n",
        "    model = Model(VOCAB_SIZE, config.EMB_DIM, HIDDEN_DIM, OUTPUT_DIM, embedding_layer)\n",
        "    model = model.to(config.DEVICE)\n",
        "    \n",
        "#     model\n",
        "#     model = Model(2, len(dataset.source_vocab), 128, 100, 1 ).to(config.DEVICE)\n",
        "#     hidden = model.init_hidden(config.BATCH_SIZE)\n",
        "#     model.hidden = hidden\n",
        "\n",
        "    loss_fn = nn.BCELoss()\n",
        "    optimizer = torch.optim.SGD(model.parameters(), lr = 0.1)\n",
        "    \n",
        "    train_losses = []\n",
        "    val_losses = []\n",
        "    train_accs = []\n",
        "    val_accs = []\n",
        "\n",
        "    print(f\"-----------------------------------------------------------{fold}-fold of the model-----------------------------------------------------------\")\n",
        "    for epoch in range(config.EPOCHS):\n",
        "        train_loss, train_correct = train_epochs(train_loader, model, loss_fn, optimizer)\n",
        "        val_loss, val_correct = val_epochs(val_loader, model, loss_fn)  \n",
        "        \n",
        "        train_loss = train_loss/len(train_loader.sampler)\n",
        "        val_loss = val_loss/len(val_loader.sampler)\n",
        "        train_acc = (train_correct/len(train_loader.sampler))*100\n",
        "        val_acc = (val_correct/len(val_loader.sampler))*100\n",
        "        \n",
        "        train_losses.append(train_loss)\n",
        "        val_losses.append(val_loss)\n",
        "        train_accs.append(train_acc.cpu().numpy().tolist())\n",
        "        val_accs.append(val_acc.cpu().numpy().tolist())\n",
        "        \n",
        "        print(f\"| Train Loss : {train_loss} |\", end = \" \")\n",
        "        print(f\" Val Loss : {val_loss} |\", end = \" \")\n",
        "        print(f\"Train Acc : {train_acc} |\", end = \" \")\n",
        "        print(f\"Val Acc : {val_acc} |\")\n",
        "\n",
        "        \n",
        "    # Saving the state dicts for the model\n",
        "    model_state_dicts.update({f\"LSTM-Model-for-{fold}\" : model.state_dict(),\n",
        "                             f\"Model-Optimizer-for-{fold}\" : optimizer.state_dict()})\n",
        "    \n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 496
        },
        "id": "XpjZFM9lMBOi",
        "outputId": "e9e2d3b6-70ca-410f-baeb-4eac9d2dd3b5"
      },
      "execution_count": 164,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Words found are : 9997\n",
            "Embedding(9997, 100)\n",
            "-----------------------------------------------------------0-fold of the model-----------------------------------------------------------\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "  4%|▍         | 299/7310 [00:14<05:40, 20.57it/s]\n"
          ]
        },
        {
          "output_type": "error",
          "ename": "IndexError",
          "evalue": "ignored",
          "traceback": [
            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
            "\u001b[0;31mIndexError\u001b[0m                                Traceback (most recent call last)",
            "\u001b[0;32m<ipython-input-164-8e0be832d3b1>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     45\u001b[0m     \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34mf\"-----------------------------------------------------------{fold}-fold of the model-----------------------------------------------------------\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     46\u001b[0m     \u001b[0;32mfor\u001b[0m \u001b[0mepoch\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mEPOCHS\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 47\u001b[0;31m         \u001b[0mtrain_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtrain_correct\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtrain_epochs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrain_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mloss_fn\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     48\u001b[0m         \u001b[0mval_loss\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mval_correct\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mval_epochs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mval_loader\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mloss_fn\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     49\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-162-ee650733b74b>\u001b[0m in \u001b[0;36mtrain_epochs\u001b[0;34m(dataloader, model, loss_fn, optimizer)\u001b[0m\n\u001b[1;32m      9\u001b[0m         \u001b[0mreview\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mreview\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDEVICE\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDEVICE\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     10\u001b[0m         \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzero_grad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 11\u001b[0;31m         \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mreview\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     12\u001b[0m         \u001b[0moutput\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moutput\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreshape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     13\u001b[0m         \u001b[0mloss\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mloss_fn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0moutput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlabel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1128\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1129\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1130\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1131\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1132\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m<ipython-input-161-ab915fbbbd65>\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, text)\u001b[0m\n\u001b[1;32m     22\u001b[0m         \u001b[0mhidden\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mhidden\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDEVICE\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     23\u001b[0m         \u001b[0mmemory\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmemory\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mconfig\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mDEVICE\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 24\u001b[0;31m         \u001b[0membedded\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0membedding\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtext\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     25\u001b[0m         \u001b[0moutput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhidden\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlstm\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0membedded\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mhidden\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmemory\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     26\u001b[0m \u001b[0;31m#         assert torch.equal(output[-1,:,:], hidden.squeeze(0))\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py\u001b[0m in \u001b[0;36m_call_impl\u001b[0;34m(self, *input, **kwargs)\u001b[0m\n\u001b[1;32m   1128\u001b[0m         if not (self._backward_hooks or self._forward_hooks or self._forward_pre_hooks or _global_backward_hooks\n\u001b[1;32m   1129\u001b[0m                 or _global_forward_hooks or _global_forward_pre_hooks):\n\u001b[0;32m-> 1130\u001b[0;31m             \u001b[0;32mreturn\u001b[0m \u001b[0mforward_call\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1131\u001b[0m         \u001b[0;31m# Do not call functions when jit is used\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1132\u001b[0m         \u001b[0mfull_backward_hooks\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnon_full_backward_hooks\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/modules/sparse.py\u001b[0m in \u001b[0;36mforward\u001b[0;34m(self, input)\u001b[0m\n\u001b[1;32m    158\u001b[0m         return F.embedding(\n\u001b[1;32m    159\u001b[0m             \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mpadding_idx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmax_norm\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 160\u001b[0;31m             self.norm_type, self.scale_grad_by_freq, self.sparse)\n\u001b[0m\u001b[1;32m    161\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    162\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mextra_repr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0mstr\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;32m/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py\u001b[0m in \u001b[0;36membedding\u001b[0;34m(input, weight, padding_idx, max_norm, norm_type, scale_grad_by_freq, sparse)\u001b[0m\n\u001b[1;32m   2197\u001b[0m         \u001b[0;31m# remove once script supports set_grad_enabled\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2198\u001b[0m         \u001b[0m_no_grad_embedding_renorm_\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmax_norm\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnorm_type\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 2199\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0mtorch\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0membedding\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mweight\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpadding_idx\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mscale_grad_by_freq\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msparse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   2200\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   2201\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
            "\u001b[0;31mIndexError\u001b[0m: index out of range in self"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "torch.save(model_state_dicts, \"My-Model.pt\")"
      ],
      "metadata": {
        "id": "_9SLBmihMBLu"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Inference"
      ],
      "metadata": {
        "id": "sYB-OC_jMNc9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "def numericalize(text):\n",
        "    \n",
        "    numerialized_source = [] \n",
        "    numerialized_source = [dataset.source_vocab.stoi[\"start\"]]\n",
        "    numerialized_source += dataset.source_vocab.numericalize(text)\n",
        "    numerialized_source.append(dataset.source_vocab.stoi[\"end\"])\n",
        "    \n",
        "    return numerialized_source\n",
        "\n",
        "def padding(source):\n",
        "    padded_sequence = torch.zeros(config.MAX_LEN, 1, dtype = torch.int)\n",
        "    source = torch.tensor(source)\n",
        "    \n",
        "    if len(source) > config.MAX_LEN:\n",
        "        padded_sequence[:, 0] = source[: config.MAX_LEN]\n",
        "    else:\n",
        "        padded_sequence[:len(source), 0] = padded_sequence[:len(source), 0] + source\n",
        "    \n",
        "    return padded_sequence"
      ],
      "metadata": {
        "id": "pUIoxY0mMBHr"
      },
      "execution_count": 171,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "def infer_processing(text):\n",
        "    \n",
        "    text = preprocessing(text)\n",
        "    text = numericalize(text)\n",
        "    text = padding(text)\n",
        "    return text"
      ],
      "metadata": {
        "id": "aTdbhoDAMBEm"
      },
      "execution_count": 166,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "aspects = [\"food\", \"service\", \"place\", \"price\"]\n",
        "\n",
        "def get_similarity(text, aspect):\n",
        "    try:\n",
        "#         text = \" \".join(text)\n",
        "        return fasttext_model.wv.similarity(text, aspect)\n",
        "    except:\n",
        "        return 0\n",
        "    \n",
        "def best_aspect(text, aspects):\n",
        "    a = []\n",
        "    \n",
        "    for aspect in aspects:\n",
        "        a.append(get_similarity(text, aspect))\n",
        "    \n",
        "    return aspects[np.argmax(a)]"
      ],
      "metadata": {
        "id": "eANW5tWpMBB-"
      },
      "execution_count": 167,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "sample = \"Love going here for happy hour!\"\n",
        "\n",
        "ba = best_aspect(preprocessing(sample), aspects)\n",
        "\n",
        "a = infer_processing(sample).to(config.DEVICE)"
      ],
      "metadata": {
        "id": "Z79CXAVwMA_l"
      },
      "execution_count": 200,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "model.eval()\n",
        "sentiment = model(a)\n",
        "sentiment = sentiment.cpu().detach().numpy()[0]\n",
        "\n",
        "if sentiment > 0.3:\n",
        "    sentiment = 'Positively'\n",
        "else :\n",
        "    sentiment = 'Negatively'"
      ],
      "metadata": {
        "id": "dYrpZxNFMA82"
      },
      "execution_count": 201,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"The reviewer is talking {sentiment} about the {ba} of the restaurant in his/her comment\")"
      ],
      "metadata": {
        "id": "MqyCi3C9MA6Q",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "1f89c9ed-8911-4ce3-acb2-e2cb7cce930e"
      },
      "execution_count": 202,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The reviewer is talking Positively about the place of the restaurant in his/her comment\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "J3TmBWlyMAwq"
      },
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.7.3"
    },
    "colab": {
      "provenance": [],
      "collapsed_sections": [],
      "include_colab_link": true
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}