Early Stroke Prediction Using Machine Learning

Abstract

Stroke is one of the most severe diseases globally, and it is directly or indirectly responsible for a considerable number of deaths. A variety of data mining techniques are employed in the health care industry to aid in diagnosing and early detection of illnesses. Several elements that lead to stroke are considered in the current investigation. First, we're looking into the characteristics of those who are more likely to suffer from a stroke than others. The dataset is obtained from a freely available source, and multiple classification algorithms are used to predict the occurrence of a stroke shortly. By employing the random forest algorithm, it has been possible to obtain an accuracy of 98.94 percent. Finally, various preventative steps such as quitting smoking, avoiding alcohol, and other factors are recommended to reduce the risk of having a stroke.