SF Airbnb pricing model

Predicting Airbnb prices in San Francisco using machine learning regression models.

Features:

• Outlier removal to filter luxury listings above $500
• One-hot encoding for categorical features (neighbourhood, room type)
• Trained on 4,477 listings after outlier removal.
• Hyperparameter tuning with 5-fold GridSearchCV across all three models: Decision Tree Regressor, Random Forest Regressor, and XGBoost Regressor
• Model comparison using MAE and R² Score

Technologies:

• scikit-learn
• pandas
• XGBoost
• Python
• Jupyter Notebook

How to run:

• Clone the repository
• Install dependencies: pip install pandas scikit-learn xgboost jupyter
• Prepare dataset: listingsSF.csv (from Inside Airbnb)
• Open SFAirbnbPricing.ipynb in Jupyter and run