Enhancing Accuracy in Predicting Continuous Values through Regression

Abstract

Enhancing the accuracy of predicting continuous values is essential in many fields. Regression is a practical approach in data mining, and machine learning can achieve this task. This study proposes a new framework of multiple regression models to obtain high accuracy using the Boston House Pricing Dataset (BHD). The examined models involve simple linear, multiple linear, Polynomial, Lasso, Ridge, Random Forest, Keras, and Gradient Boosting regression to seek a fair comparison with the best experimental result. The attempt is to select the best-predicting model using evaluation indicators such as R-squared Score (R2), Mean Squared Error (MSE) and Mean Absolute Error (MAE). Among the examined models, the first promising outcomes indicate that Random Forest and Ridge regressors scored a high level of R2 i.e. 89.9 and 88.3, respectively. In addition, The Gradient Boosting model offers the best result of R2 92 with MSE 0.72 and MAE 2.00. This research proposes two techniques to improve the accuracy of the best model. Re-sampling and optimization using the RandomizedSearchCV tuned hyper-parameter enhances the R2 score to 93.2 with a better MSE of 0.015 and MAE of 0.82. These findings prove a significant improvement in model performance and potential for practical application in real-world scenarios.