Data Interpretation and Early Stroke Prediction using Ensemble Technique

A stroke is a medical emergency that occurs when there is an obstruction or rupture in the blood vessels. In order to find the risk factors in the dataset statistical analysis is performed. The importance of statistical analysis in understanding its prevalence and impact cannot be overstated. Statistical analysis enables academics and healthcare providers to investigate stroke-related data thoroughly. It aids in identifying patterns, trends, and relationships within data that can provide useful insights into stroke risk factors, prevention measures, and treatment outcomes. The goal of this study aims to deal with the class imbalance issue where the model gets biased towards the majority class which leads to incorrect accuracy in order to this combat the SMOTE technique is used apart from that this study helps to examine the association between the stroke risk factors using the chi-square test and Pearson’s correlation. As a result, it was concluded that hypertension, heart disease, ever married, work type, and smoking status are highly correlated. The chi-square test is only applicable to categorical variables, whereas Pearson’s correlation is only applicable to continuous variables. The stroke can be anticipated early employing well-known machine learning approaches such as random forest, xgboost, and Support Vector Machine(SVM), with random forest outperforming the others.