Heart Disease Prediction in Indonesia Using Machine Learning Techniques for Multiclass Classification

This research concentrates on utilizing multiclass classification methodology for heart disease prediction. Specifically, this paper explores the utilization of Support Vector Machine (SVM) and Multi-Layer Perceptron (MLP) classifiers to classify various cardiac conditions. This research is used to get the best results in predicting heart disease that has a classification of more than two categories using machine learning methods. The main goal is to contribute insights to medical diagnostics by explaining the effectiveness of SVM and MLP in improving the accuracy of heart disease prediction models in various classes using several factors that influence heart disease. The data used in this research has missing values, therefore in this research, several methods were used to overcome missing values. Determining the best model is done by comparing precision, recall, and accuracy values. The best model in this research is the SVM model because it has the highest accuracy, precision, and recall. It is hoped that this research can provide knowledge to optimize its application in the identification and classification of heart disease to make it more efficient.