AN ENHANCED SYSTEM FOR PREDICTING HEART STROKE USING MACHINE LEARNING

Abstract Globally, heart disease and strokes are now among the top causes of death, particularly for adults. According to the WHO's data analysis, 17.9 million persons worldwide are estimated to have died from cardiovascular disease (CVD) in 2019. Heart attacks and strokes were responsible for almost 85% of all deaths. Hence, it is essential to identify cardiovascular diseases (CVD) so that the right treatments can be given as soon as feasible. Age, gender, hypertension, heart disease, type of work, type of residence, average blood glucose level, BMI, and smoking status were the parameters we collected from the Kaggle dataset, and predictions were made using machine learning algorithms like Nave Bayes, Random Forest, and Decision Tree Classifier. Many models were compared in a predictive analysis, and the Decision Theory classifier was determined to be the most accurate of them, with an accuracy of 97.6%.