PDFThe proposed article identifies factors such as property locations and characteristics that drive residential housing prices. We aim to gain insights from housing price models to use in economic development, urban planning, financial services, logistics, and industrial development. To do so, we developed a predictive model for housing prices that provides an understanding of the most important factors driving property prices. In order to do this, we leveraged data sources such as Sales Disclosure Forms (STATS Indiana), Geocoding (ArcGIS with the help of IU Polis Center), property characteristic features (Melissa Data - Intrinsic), and property location-based features (Niche.com Ratings, Google Places API, School Ratings - Extrinsic).
Keywords— House Price Prediction, Random Forest, XGBoost, LightGBM, Gradient Boosting, CatBoost, Shapley Additive Explanations