Abstract: This study investigates the complex interplay between general health parameters—most notably blood pressure and the risk of brain stroke through a data-driven approach using machine learning algorithms. By leveraging a comprehensive dataset and applying various classification models including Random Forest (RF), Decision Tree (DT) and Artificial Neural Networks (ANN), the research evaluates each model’s effectiveness in predicting stroke occurrences. Feature engineering, data pre-processing, and statistical validation techniques are employed to enhance model performance and accuracy. The study not only identifies key predictors of stroke but also offers a comparative analysis of algorithmic performance, paving the way for intelligent diagnostic systems that support early detection and preventive healthcare strategies.

Keywords: Brain stroke prediction, Machine learning, including Random Forest (RF), Decision Tree (DT) and Artificial Neural Networks (ANN), Predictive analytics, Health informatics, Feature engineering, Data pre-processing, Stroke risk assessment.

| DOI: 10.17148/IARJSET.2025.125260