查看文章

This study presents the development and comparison of four machine learning (ML) models, namely Linear Regression (LR), Decision Tree (DT), Random Forest (RF), and k-Nearest Neighbours (k-NN), for predicting house prices using the Boston Housing Dataset. The performance of these models was evaluated using metrics such as root mean square error (RMSE), and R-squared (R2), with the aim of identifying the model that best predicts housing prices. The dataset was thoroughly analyzed for features, correlations, multicollinearity, and overfitting. Results indicate that the RF model outperformed the other models in predicting house prices, due to its ability to handle non-linearity and complex interactions among variables and reduce the impact of outliers. The DT model also performed well but may have been more prone to overfitting. LR, on the other hand, may have been limited by its assumptions of linearity and independence among variables.