<![CDATA[Indonesia's rapid urbanization, particularly in the Jabodetabek region, has created a severe housing shortage with a backlog of 2.93 million units representing 30% of the national deficit. This imbalance between supply and demand necessitates accurate house price prediction systems to guide market participants. This research implements Random Forest Regression algorithm to predict house prices in the Jabodetabek region using comprehensive datasets covering land area, building area, geographical location, room quantities, facilities, and property characteristics across districts and cities. The methodology involves data preprocessing, model training using Random Forest Regression, and performance evaluation using established metrics. Results demonstrate great algorithm performance with RMSE of 0.3545, MAE of 0.2014, MAPE of 1.0184, and R² of 0.8751 confirming the model explains 87.51% of house price variance. The implementation successfully addresses the research objective of providing developers with a reliable algorithmic framework for property pricing strategies.]]>
Copyrights © 2025
