Comparative Diagnosis Model for Risk Factors of Heart Disease Using Deep Learning

Many sharing datasets about heart disease including Kaggle and UCI. Those statistics are crucial for human being to prepare themselves. This research has been conducted to create a diagnostic model for risk factors for developing heart disease. Five commonly used classification models are implemented in this paper including 1. Neural Network 2. Support Vector Machines (SVM) 3. Decision Tree Classification 4. Random Forest Classifier 5. Logistic Regression to compare the accuracy. The results of the experiment compared the performance of the models of all 5 algorithms, finding that the Neural Network model had the best performance. It has the highest accuracy and accuracy at 96.02%. It is the highest accuracy and suitable for use in diagnosing factors in heart disease.