Heart disease prediction system using (SMOTE technique) balanced dataset and decision tree classifier Available to Purchase

The deadliest disease and a major cause of the mortality worldwide is heart disease. In medical scope, Machine Learning (ML) is becoming increasingly important. In this work, the SMOTE technique balanced dataset is utilized for the improvement of the performance of the prediction of heart disease, and Cleveland Heart Disease Dataset is predicted using the Decision Tree (DT) algorithm. The dataset contains 14 key attributes that were utilized in the investigation. Yet, classes are not often balanced, and data imbalances develop in the case when one class is a minority and the other is a majority. The usage of the SMOTE resampling technique for balancing the data was examined in this research, and the outcomes of the DT algorithm were compared for unbalanced and balanced data. According to the results of the experiments, classification with resampling/balancing improves accuracy by up to 18.1%. The accuracy of DT without balanced data is 73.3%, whereas the accuracy of DT with balanced data is 91.4%.