Predict whether a patient is likely to get a stroke based on demographic and medical history information.
This dataset comes from Kaggle. It containing demographic and medical history information on individuals along with an indicator of whether they experienced a stroke.
Three predictive models were constructed: two logistic regression models and one decision tree. The first logistic regression model incorporated heart disease, hypertension, and smoking status as covariates, while the second model expanded upon these variables to include age, average glucose level, and BMI. Additionally, a decision tree model was built to predict stroke probability based on a combination of these features.