Abstract :

Predicting housing prices accurately is a vital problem in the real estate industry, assisting buyers, sellers, investors, and policymakers in making informed decisions. This project presents a web-based predictive system that applies regression-based machine learning models to estimate housing prices using key parameters such as location, size, number of rooms, neighborhood facilities, and market conditions. A full-stack architecture integrates a trained ML model with an interactive user interface and efficient backend processing. The system allows users to input property attributes and returns real-time price predictions. It emphasizes usability, scalability, and reliable data-driven insights. Performance evaluation metrics such as RMSE, MAE, and R² are used to validate model efficiency. The project demonstrates the feasibility of combining artificial intelligence with web technologies to build an intelligent real estate decision support application.