Diagnosis of heart disease using oversampling methods and decision tree classifier in cardiology

Heart disease is one of the most prevalent and critical diseases that endangers lives human beings. In addition to clinical diagnosis, machine learning deep learning-based approaches are vital in diagnosis heart disease. This paper proposes a balanced optimized machine-learning algorithm for detection. technique combines oversampling techniques, attribute pruning, CART decision tree classifier, rule pruning through hyper-parameter tuning identify presence It further identifies key attributes contribute occurrence malfunctioning. Experimental results show SMOTE sampled dataset exhibits effective performance when implemented using algorithm, with an improvement 11%, 75%, 62%, 71% accuracy, precision, recall, f1 scores compared was not subjected sampling. The works effectively imbalance ratio high dataset. can be used predict even highly imbalanced datasets features malfunctioning heart.