California Housing Affordability Optimization

An ML- and optimization-driven decision-support system for housing affordability in California, framing housing selection as a multi-objective optimization problem over affordability, safety, and accessibility rather than a traditional recommender system.

The project combines predictive modeling (Random Forest, Gradient Boosting) with constraint-based optimization (MILP and heuristic methods) to maximize Quality of Life (QoL) under real-world affordability and policy constraints, enabling scenario analysis for individual and policy-level evaluation.