Application of Feature Engineering Techniques and Machine Learning Algorithms for Property Price Prediction

Sihombing, Denny Jean Cross

JITSI : Jurnal Ilmiah Teknologi Sistem Informasi
Vol 5 No 2 (2024)

This research applies feature engineering techniques and machine learning algorithms to predict property prices using a dataset from Kaggle. Three models were implemented: Linear Regression, Decision Tree, and Random Forest. The Random Forest model demonstrated the best performance with an average Mean Absolute Error (MAE) of 16472.76, Mean Squared Error (MSE) of 457407807.78, and R-squared (R²) of 0.83. Key features influencing property prices were identified through feature importance analysis, providing valuable insights for enhancing property appraisals and investment decisions.