Research Article
Boston Housing Price Prediction and Factor Analysis: A Comparative Study of Linear Regression and Random Forest
@INPROCEEDINGS{10.4108/eai.22-5-2026.2365118, author={Jixuan Chen}, title={Boston Housing Price Prediction and Factor Analysis: A Comparative Study of Linear Regression and Random Forest}, proceedings={Proceedings of the 4th International Conference on Image, Algorithms, and Artificial Intelligence, ICIAAI 2026, 22-24 May 2026, Singapore, Singapore}, publisher={EAI}, proceedings_a={ICIAAI}, year={2026}, month={8}, keywords={Boston Housing Price Prediction Linear Regression Random Forest Feature Selection}, doi={10.4108/eai.22-5-2026.2365118} }- Jixuan Chen
Year: 2026
Boston Housing Price Prediction and Factor Analysis: A Comparative Study of Linear Regression and Random Forest
ICIAAI
EAI
DOI: 10.4108/eai.22-5-2026.2365118
Abstract
Accurate housing price prediction is important for real estate transactions, urban planning, and policy making. This study uses Boston housing data to find key influencing factors through correlation analysis and significance testing. It builds linear regression and random forest models, and evaluates their performance using R², MAE, and RMSE. The results show that the random forest model has better fitting accuracy and generalization ability. Its test set R² is 0.89, which is 15% higher than the linear regression model. Finally, the paper summarizes the findings, discusses limitations, and suggests future work such as integrating multi-source data. This provides practical references for housing price prediction in similar urban contexts.
