Exploring the Impact of Univariate Feature Selection Method on Machine Learning Algorithms for Heart Disease Prediction

This study compares various machine learning techniques such as logistic regression, support vector machine (SVM), random forest, Naïve Bayes, and decision tree. Later, a feature selection approach that considers the Univariate Feature Selection (UFS) method was applied to select the best features based on statistical testing. In this research, we employed the ANOVA F-test to pick the most significant features, and we used a cutting-edge dataset for experimental analysis. The ANOVA F-test was used to exploit the target variable's dependency on the continuous variables. The results show that employing the UFS approach for picking features improves the performance of the algorithms. According to the experimental results, using UFS with the logistic regression technique yields 93.3% accuracy, whereas the standalone Naïve Bayes algorithm achieved the best accuracy.