Elimination of Irrelevant Features and Heart Disease Recognition by Employing Machine Learning Algorithms using Clinical Data

A heart disease diagnosis method has been proposed for effective heart disease diagnosis. In the proposed method Machine Learning (ML) classifiers have been used for detection of heart disease. Chi square feature selection algorithm has been used for related feature selection to improve the prediction performance of machine learning models. Cross validation, method Hold out has been employed for model hyper parameters tuning and best model selection. Furthermore, performance evaluation metrics, such as classification accuracy, specificity, sensitivity, Matthews' correlation coefficient and execution time have been used for model performance evaluation. The Cleveland heart disease data set has been used for testing of the proposed method. The experimental results demonstrated that proposed method has achieved high performance as compared to state of the art methods. Furthermore, the proposed method performance has been compared with deep learning model. Thus, the proposed method will support the medical professional to diagnosis heart disease efficiently and could easily incorporated in healthcare for diagnosis of heart disease.