Heart disease prediction using ensemble method for feature selection

Abstract

Data mining (DM) is a scheme in which useful information is extracted from the unstructured data. On the basis of existing information, the prediction analysis (PA) method is used to forecast future possibilities. This investigate work is arranged on the premise of foreseeing the cardiac illness. To pre-process the data, extract the attributes, and classify the data, are all steps in forecasting coronary artery disease. This work projected a hybrid model in which Random Forest ensemble technique is put together with Logistic Regression. The features are extracted using RF ensemble method and the LR algorithm is assisted in classifying the disease. Different metrics are utilized to quantify the projected model. The evaluation revealed that the accuracy of the projected model is 95% to predict the cardiac disease.