Early prediction of heart disease by using machine learning algorithm

Abstract

It is critical to diagnose and treat heart disease early because it is a leading cause of death worldwide. The writers of this piece zeroed in on the efficiency of a machine learning algorithm for the early diagnosis of cardiac illness. Applying a Kaggle dataset, we create and evaluate two ML models: Support Vector Machine (SVM) and K-Nearest Neighbors (KNN). Finding the most accurate models in terms of projected outcomes and discussing their consequences is the main goal of this study. Feature extraction and splitting are two of the many operations performed on the dataset to guarantee efficient model training. The findings point to the possibility that machine learning might greatly improve the accuracy of heart disease predictions, which would be good for early disease detection since it would aid in identifying of the diseases. Both the SVM and KNN algorithms prove to be effective in the present work, and observe equal to 90. 2% as well as 86% accuracy correspondingly.