Stroke Prediction Based on Random Forest with SMOTE

A stroke happens when the brain's blood flow is interrupted or diminished as a result of a blood vessel rupture (hemorrhagic stroke) or blockage (ischemic stroke). The absence of blood supply prevents the brain from receiving oxygen and nutrients, which causes some brain cells to perish. This disorder impairs the proper operation of bodily functions that are regulated by injured brain regions. It is the leading cause of death and disability globally. Most of research has focused on predicting heart attacks, but relatively little on the danger of brain attacks. In light of this, numerous machine learning models were developed to forecast the probability of a brain stroke. This paper used a variety of physiological factors and machine learning algorithms like Logistic Regression, Decision Tree Classification, Random Forest Classification, K-Nearest Neighbors, Naive Bayes, and Support Vector Machine Classification to train five different models for accurate prediction. This study shows that the Random Forest algorithm has the best performance compared to other algorithms with an accuracy of 82.07%.