Stroke is a serious medical condition that affects millions of people worldwide. Early detection and prediction of stroke can help in better management and prevention of the condition. In this study, we compare the performance of five different machine learning algorithms for stroke prediction. The algorithms tested were Gaussian Naive Bayes, logistic regression, decision tree classifier, K-nearest neighbors classifier, and gradient boosting classifier. We used a dataset of stroke patients with various features such as age, hypertension, heart disease, and smoking status. The dataset was preprocessed and cleaned before training the models. The models were evaluated using k-fold cross-validation and their accuracy scores were recorded. Our results show that logistic regression, K-nearest neighbors classifier, and gradient boosting classifier achieved the highest accuracy scores, with values of 98.49%, 98.48%, and 98.44%, respectively. Decision tree classifier achieved an accuracy score of 96.95%, while Gaussian Naive Bayes achieved an accuracy score of 16.81%. Our findings suggest that ML algorithms such as LR, KNN Classifier, and gradient boosting classifier can be effective in predicting stroke. These models can be used by healthcare professionals to identify patients at high risk of stroke and provide timely interventions to prevent adverse conditions (Koh and Tan, J Healthc Inf Manage 19(2):65, 2011).