Emphasised Research on Heart Disease Divination Applying Tree Based Algorithms and Feature Selection

Abstract

Heart disease has evolved to become the most deadly ailment on the earth, and it has been the top reason for mortality worldwide. As a result, a dependable, efficient, and practical method for diagnosing and treating such disorders promptly is required. This study examines and compares several Machine Learning (ML) algorithms and approaches. Six ML classifiers are tested to see which one's the most successful at diagnosing heart disease. Tree-based techniques are among the most basic and extensively used ensemble learning approaches. According to the analysis, tree-based models such as Decision Tree (DT) and Random Forest (RF) deliver actionable insights with high efficacy, uniformity, and applicability. Relevant features are identified by using the Feature Selection (FS) process, and the output of classifiers is calculated based on these features. FS removes irrelevant features without impacting learning output. Our research intends to improve the system's efficiency. The goal of this research is to combine FS with tree-based algorithms to improve the accuracy of heart disease prediction.