Skip to content

The repo is created to show case the sentiment Analysis performed IMDB dataset using the State of the Art Classification approaches.

License

Notifications You must be signed in to change notification settings

chaitanya2593/IMDB_Sentiment_Analysis

Repository files navigation

IMDB Reviews Sentiment Analysis

Table of Contents
  1. About The Project
  2. Getting Started
  3. Code Usage
  4. License
  5. Contact
  6. Reference

About The Project

  • The task of classifying sentiments of texts (for example movie or product reviews) has high practical significance in online marketing as well as financial prediction. This is a non-trivial task, since the concept of sentiment is not easily captured.
  • As part of the model building five different approaches where used and please go through the notebooks for more understanding.

Product Name Screen Shot

(back to top)

Built With

  • Python
    • Data science packages
    • tensorflow
  • Google Colab - (Please use colab as it has already most of the packages installed)

(back to top)

Getting Started

  • Code development is performed in the Google colab notebooks. Colab already has required data science packages preinstalled and provided the pip (python package installer) commands where it is required.

  • For few data I/O operations like saving the processed files and models, I have used my personal drive.

Prerequisites

  • The user should have an google account and access to the google drive as well.
  • Kindly install the packages stated in the notebooks.

Code Usage

All the Approaches are summarized in 2 notebooks.

  • 1_Multi_models_IMDB_Sentiment_classification.ipynb

    1. TFIDF + classical statistical model
      • Random Forest
      • Naive Bias
    2. LSTM classification model
      • Untuned
      • Tuning using the Keras-Tuning
    3. LSTM model, where the embeddings are initialized with pre-trained word vectors
      • GLOVE Embedding - Wiki 2014 dump
      • GLOVE Embedding - Twitter Data
    4. fastText model - From facebook
  • 2_Transformer_Bert_IMDB_Sentiment_classification.ipynb

    • In this script, BERT pretained transformer has been used for training.

(back to top)

License

Distributed under the GNU License. See LICENSE.txt for more information.

(back to top)

Contact

  1. VS Chaitanya Madduri | LinkedIn

Project Link: https://github.com/chaitanya2593/IMDB_Sentiment_Analysis.git

(back to top)

Reference

(back to top)

About

The repo is created to show case the sentiment Analysis performed IMDB dataset using the State of the Art Classification approaches.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published