Main Article Content
Among many other diseases, heart disease is a leading ailment of today, and it affects many people all over the world. As a result, early detection and diagnosis of this disease is an essential factor in health care field. The health sector benefits greatly from an effective and reliable diagnosis of this disease. Centred on various machine learning techniques, we formulated an accurate and efficient recognition method for diagnosing heart disease in this paper. Linear Regression, Support vector machine, Linear SVC, MLP Classifier, Stochastic Gradient, Decision Tree Classifier, Random forest classifier, XGB Classifier, LGBM Classifier, Gradient Boosting Classifier, Ridge Classifier, Bagging Classifier, Extra Trees Classifier, AdaBoost Classifier, Logistic Regression, K-nearest neighbour (KNN), Naive Bayes, Neural network(NN) with Keras, Gaussian Process Classification and Voting classifier are some of the classification algorithms used in this machine learning system, while, on the other hand, for eliminating obsolete and redundant features, standard feature selection techniques like Recursive Feature Elimination (RFE),Feature Selection with the Pearson Correlation and Chi-squared method have been introduced. Feature selection algorithms are employed to enhance classification performance and minimize classification system execution time. Of all the classifiers used, Stochastic Gradient Descent classifier performed well and obtained an accuracy of 93.44%. Furthermore, the proposed scheme can be effectively applied in the health sector to correctly assess the HD.