Performance Evaluation of Machine Learning Algorithms in Design and Development of Heart Disease Detection

Abstract

Heart disease is becoming the top cause of mortality worldwide. It can be avoided by early detection and timely treatment. Machine learning (ML) and data mining techniques play an important role in the detection of heart disease. So, the current article targets the performance evaluation of various ML algorithms on heart disease detection tasks. In this context, the article has used different machine learning algorithms like Logistic Regression (LR), K-Nearest Neighbors Classifier (KNN), Support Vector Machine (SVM), Decision Tree Classifier (DT), Random Forest Classifier (RF) and eXtreme Gradient Boost Classifier (XGBoost) with standard parameter settings and Hyperparameter Tuning. All algorithms are applied to the merged Dataset (Cleveland Heart Disease Dataset + Heart Disease Dataset) and evaluation is done on the basis of accuracy. Finally, we have observed that RF is having more classification accuracy as compared to other algorithms i.e. 97.22 (in percent).