Machine Learning
House Price Prediction
ML regression system predicting house prices with R² of 0.83 using 6 models and feature engineering.
Project Overview
A real estate price prediction system trained on California Housing dataset. Engineers 4 new features from existing data, trains 6 regression models, and produces geographic price heatmap visualization.
⚠ The Problem
Real estate agents need accurate pricing tools to help buyers and sellers make informed decisions based on property features and location.
✓ The Solution
Feature engineering created rooms_per_household, bedrooms_per_room, population density, and income ratios. XGBoost achieved R² 0.83 with geographic visualization.
Results & Outcomes
Best R² 0.83 with XGBoost. Interactive geographic price heatmap. Custom property price predictor for any input combination.