Application for Early Heart Disease Prediction Based on Data Mining Approach

The purpose of this study was to develop an application for early heart disease prediction using a data mining approach. A clinical dataset of 303 patients was gathered from the UCI repository that was available at http://archive.ics.uci.edu/ml/datasets/Heart+Disease. There were 13 attributes that were used to classify the patient into two class predictions namely No presence or Have heart disease. Machine learning with CART algorithm used for backend. Flask micro web framework used for middleware and Bootstrap web framework used for frontend. The experiment result performed the application for early heart disease prediction with CART achieved performance for accuracy 88.33%, precision 88.00%, F Measure 86.28%, recall 84.62%.