• Kashish Kalpi
• Rajneesh Pachouri
• Anurag Jain
• Monali Sahoo

Stroke is a critical medical condition requiring swift and accurate diagnosis to minimize long-term consequences. This project introduces an advanced machine learning-based stroke diagnosis system leveraging the Random Forest Classifier and Bagging Classifier. Comprehensive preprocessing, including handling missing values, normalizing data, and addressing imbalances through techniques like resampling, ensures the dataset's quality. The Random Forest Classifier achieved a train accuracy of 100% and test accuracy of 99%, while the Bagging Classifier attained train and test accuracies of 99% and 98%, respectively. By analyzing key patient attributes such as age, hypertension, heart disease history, glucose level, and BMI, the models demonstrated exceptional predictive capability. This system supports early and reliable stroke detection, emphasizing model interpretability, ethical considerations, and real-world applicability in clinical settings. The findings underscore the transformative potential of machine learning in enhancing stroke diagnosis and patient care.

Stroke Diagnosis, Random Forest Classifier, Bagging Classifier,Machine Learning , Data Preprocessing

"Enhanced Stroke Diagnosis Prediction System Using Random Forest Classifier compared with Bagging Classifier", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 1, page no.f53-f58, January-2025, Available :http://www.jetir.org/papers/JETIR2501507.pdf

"Enhanced Stroke Diagnosis Prediction System Using Random Forest Classifier compared with Bagging Classifier", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 1, page no. ppf53-f58, January-2025, Available at : http://www.jetir.org/papers/JETIR2501507.pdf