Skip to content

Latest commit

 

History

History
73 lines (49 loc) · 2.81 KB

README.md

File metadata and controls

73 lines (49 loc) · 2.81 KB

🏠 Airbnb Price Prediction Project

Analyzing and Predicting Airbnb Listing Prices

📖 Project Overview

This project focuses on exploring and predicting the prices of Airbnb listings using factors like location, property type, availability, and reviews. Through Exploratory Data Analysis (EDA) and machine learning techniques, the goal was to uncover insights about price determinants and create a predictive model to estimate listing prices.

🚀 Key Highlights

1. Data Cleaning and Preprocessing

  • Handled missing values in key features like reviews_per_month and availability_365.
  • Addressed outliers in price and other numerical columns to improve data quality.
  • Converted categorical features, such as neighborhood and room_type, into numerical representations for model compatibility.

📊 Key Insights from EDA

1. Neighborhood Influences Price

Listings in popular neighborhoods tend to have higher average prices. Proximity to central locations and tourist spots significantly impacts pricing.

2. Seasonality in Pricing

Prices exhibit seasonal trends, with peaks during holidays and vacation periods.

3. Impact of Reviews

Properties with more reviews and higher ratings tend to be priced higher, indicating a correlation between customer trust and price.

🧠 Modeling Process

1. Data Preparation

  • Standardized numerical features, such as price and minimum_nights.
  • Encoded categorical variables for use in machine learning models.

2. Model Training and Evaluation

  • Models tested:

    • Linear Regression: A baseline model for interpretability.
    • Random Forest Regressor: Captured non-linear relationships effectively.

  • Best Model:
    The Random Forest Regressor achieved the best performance with the following metrics:

    • R² Score: 0.78
    • Mean Absolute Error (MAE): $30.20

🔑 Results and Insights

  • Neighborhood and property type emerged as key predictors of listing price.
  • The machine learning model provides a foundation for Airbnb hosts to estimate competitive prices.

🛠️ Technologies Used

  • Programming Language: Python
  • Libraries: Pandas, NumPy, Seaborn, Matplotlib, Scikit-learn

🎯 Challenges Encountered

  1. Data Quality Issues: Missing values and outliers required extensive cleaning.
  2. Feature Selection: Balancing model complexity with the interpretability of features.

📈 Next Steps

  • Expand the dataset to include listings from additional cities for broader insights.
  • Explore advanced modeling techniques to improve prediction accuracy.
  • Develop visual dashboards for better user interaction with the model results.