House Price Prediction Analysis Using Linear Regression and Random Forest Algorithms

Abstract

This study aims to analyze house price prediction using two machine learning algorithms: Linear Regression and Random Forest. Quantitative evaluation is conducted using four main metrics: Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), R² Score, and Mean Absolute Percentage Error (MAPE). The experimental results show that the Random Forest model outperforms Linear Regression in all four evaluation metrics. The MAE and RMSE of the Random Forest model are lower, indicating that this model is more effective in minimizing prediction errors. Additionally, the higher R² Score demonstrates the model's better ability to explain house price variance, while the smaller MAPE indicates more accurate prediction errors in the context of real estate. These findings suggest that choosing the right algorithm is crucial for modeling complex house price data, and although Random Forest is more accurate, its black-box nature limits interpretability. Therefore, for future research, more interpretable methods such as XGBoost with SHAP analysis could be considered.