House Value Estimation using Different Regression Machine Learning Techniques

This study investigates the effectiveness of various regression algorithms in estimating house values using a dataset sourced from Zillow.com, encompassing 15,000 residential properties from Denver, Colorado. Comparisons of different models such as linear regression, Ridge regression, Lasso regression, Elastic Net, Decision Tree, Random Forest, Gradient Boosting, and XGBoost. The models were evaluated using R-squared (R2) and Mean Absolute Error (MAE) as performance metrics. The results demonstrated that the Random Forest Regressor and XGB Regressor outperformed other models, achieving the highest R2 scores and the lowest MAE values. These findings underscore the potential of these models for accurate house price estimation, which can be instrumental for the real estate market. Accurate valuations can help prevent overpricing, which causes properties to remain unsold for extended periods, and under-pricing, leading to financial losses. Implementing these regression models can enhance pricing strategies, ensuring efficient buying and selling processes and contributing to the overall financial health of the real estate market. Future research will explore the use of a broader range of regression models with fewer features to assess their performance and robustness in house price prediction.