Heart Disease Classification Based on Personal Indicators

Abstract

This work focuses on implementing an artificial neural network to help diagnose heart diseases based on some personal measurements, such as BMI, Diabetic, and skin cancer. Oversampling and undersampling are adopted to balance the imbalanced dataset to improve the performance of the model by eliminating the bias. The recall for diagnosing heart diseases after applying oversampling and undersampling to the dataset reached 82% and 83%, respectively.