Heart Stroke Prediction Using Machine Learning: A Comparative Analysis and Implementation

One of the most challenging tasks in the medical field these days is predicting heart attacks. A heart stroke claims the life of one person every minute in the modern era. Processing vast amounts of data in the healthcare industry requires the use of data science. Given how difficult it is to predict heart attacks, the prediction process must be automated in order to minimize risks and notify patients well in advance. In this paper, the heart stroke dataset is used. The suggested work uses various data mining techniques, including SVM, Neural Network and Naïve Bayes to predict the likelihood of a heart attack and categorize the patient's risk level. This paper analyzes the performance of various machine learning algorithms in order to present a comparative study. When compared to other ML algorithms that were used, the trial results confirm that the Naïve Bayes algorithm had the highest accuracy, at 96.30 percent.