Abstract This paper introduces a comprehensive algorithmic model for predicting house prices, addressing the absence of a standard reference for property valuation. To counter this, we utilize a set of machine learning techniques that consider various house attributes and features, thereby providing a more standardized approach to house pricing. The dataset used in this study is obtained from Kaggle. A range of algorithms, including Gradient Boost, Support Vector Regression, Decision Tree, Random Forest, Bagging Tree, Ridge, Lasso, Elastic Net, and Stacking Regression, are applied to improve prediction accuracy. Stacking Regression demonstrates the potential for achieving superior prediction scores compared to conventional algorithms. Our experimental results reveal that a Stacking Regression model incorporating Gradient Boosting, Bagging Tree, and Ridge as input algorithms outperforms the other methods, yielding a lower RMSE score. After parameter tuning, the best RMSE score attained with general algorithms was 0.11157 using the Ridge algorithm. In contrast, the Stacking Regression model delivered the best RMSE score of 0.10954, highlighting its enhanced predictive capabilities for house prices.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|