House Price Prediction System

Designed and implemented a machine learning–driven predictive system to estimate residential property values using socio-economic and geographic indicators from the California Housing Dataset. The project involved comprehensive data preprocessing, correlation analysis, and feature evaluation to identify key price drivers, including median income, population density, and housing age. An advanced gradient-boosting regression model was trained and evaluated to deliver accurate price predictions, demonstrating strong generalization performance across unseen data sets. The solution highlights how predictive modelling can support real estate valuation, investment analysis, and strategic housing decisions.