Exploratory Analysis on Effect of Poor Feature Selection on the Predictive Performance of Machine Learning Models

DOI

https://www.doi.org/10.59256/ijrtmr.20260604018

Published In

Volume 6, Issue 4 (2026)

Published Online

July-August 2026

Pages

165-167

Article Type

Original Article

Check for updates

Feature selection plays a crucial role in determining the predictive performance of machine learning models. While previous studies have primarily focused on developing effective feature selection techniques, the impact of poor feature selection has received comparatively less attention. This exploratory study investigates the effect of poor feature selection on the predictive performance of machine learning models. Experiments were conducted on multiple publicly available regression datasets using Linear Regression, Random Forest Regression, and Artificial Neural Networks. Model performance was evaluated using Root Mean Squared Error (RMSE), and a one-sample, one-tailed t-test was performed to assess the statistical significance of the results. The findings indicate that poor feature selection significantly degrades the predictive performance of all three models. These findings highlight the importance of feature selection in developing reliable and accurate machine learning models