Diagnosis of Heart Disease Patients Using an Ensemble Approach Followed by Variance Ranking Feature Selection Algorithm

Abstract

Heart disease is one of the most complicated diseases, and it affects a large number of individuals throughout the world. In healthcare, particularly cardiology, early and accurate detection of cardiac disease is critical. newlineA crucial data preparation job for any data mining purpose is feature selection. Making the optimum feature selection decision for a given context is challenging and time- consuming. This problem can be solved by ensemble learning. The foundation of ensemble methods is the idea that, in many cases, the combined knowledge of a group of experts with average expertise might be superior to that of a single expert with extensive knowledge. newlineThe goal of the current work is to provide a heterogeneous ensemble feature selection for categorizing heart disease. newlineThe findings of five univariate filter feature selection procedures were combined using many aggregation methods to create the proposed ensembles. Four classifiers and six heart disease datasets were used to evaluate the effectiveness of the suggested approaches. Empirical research revealed that implementing ensemble feature. newlineIn this study, imperialist competitive algorithm with meta-heuristic approach is suggested in order to select prominent features of the heart disease. This algorithm can provide a more optimal response for feature selection toward genetic in compare with other optimization algorithms. newlineThe suggested diagnosis system achieved better accuracy than previously proposed methods and can easily be implemented in healthcare to identify heart disease.