IDMS: An Integrated Decision Making System for Heart Disease Prediction

Abstract

Heart Disease, one of the deadliest human diseases worldwide, should be properly diagnosed in time and treatments should be carried out accordingly. To predict Heart Diseases, decision making systems based on classification techniques have been widely proposed in various studies. In this paper, an Integrated Decision Making System (IDMS) has been introduced for prediction of heart disease. In addition, it uses Principal Component Analysis (PCA) for dimensionality reduction, Agglomerative hierarchical clustering technique for clustering and Random Forest (RF) for classification purpose. Then, the results are compared with other six conventional classification techniques. Some experiments are performed using Cleveland Heart Disease Dataset (CHDD) sourced from UCI-ML repository and Python language concluding that the proposed system provides better results comparing with other conventional methods. The proposed integrated decision making system will help out the doctors to diagnose the heart patients professionally and it may be useful for further investigation and predictions using different datasets and resulting valuable knowledge on Heart Disease.