An Efficient Hybrid Classifier for Prognosing Cardiac Disease

Abstract

Machine learning (ML) is a powerful tool which empowers the practitioners for predictions upon any existing or real- time data. Here, the Machine first understands the valuable patterns from the dataset and then uses that information to make predictions on the unknown data. Further, classification is the commonly used machine learning approach (ML-Approach) to make such predictions. The objective of this work aims to design and development of an ensemble classifier for prognosing cardiovascular disease (heart disease). The developed classifier integrates Support Vector Machine (SVM), K–Nearest Neighbor (K-NN), and Weighted K-NN. The applicability of ensemble classifier is evaluated on the Cleveland Heart disease dataset. Some other classifiers such as Logistic Regression (LR), Sequential Minimal Optimization (SMO), K-NN+Weighted K-NN are also implemented on the same dataset to make the performance analysis. The results of this study depict the significant improvement in the Sensitivity and Specificity parameter.