This paper explores the application of machine learning in predicting obesity, a significant global health concern. We specifically examine the impact of three feature selection methods — InfoGain, Chi-squared, and ReliefF, on the performance of classification models using Random Forest and Logistic Regression algorithms. By analyzing an obesity dataset categorized into three and seven classes, we identify key features that contribute to model accuracy. The models are evaluated using several metrics: Accuracy, Precision, Recall, Specificity, Sensitivity, and Balanced Accuracy. The findings highlight the role of feature selection in model performance, with the Random Forest algorithm achieving the highest accuracy rate of 96.7%.
feature selection, machine learning, classification algorithms, obesity