Analysing an imbalanced stroke prediction dataset using machine learning techniques

A stroke is a medical condition characterized by the rupture of blood vessels within the brain which can lead to brain damage. various symptoms may be exhibited when the brain's supply of blood and essential nutrients is disrupted. To forecast the possibility of brain stroke occurring at an early stage using Machine Learning and Deep Learning is the main objective of this study. Timely detection of the various warning signs of a stroke can significantly reduce its severity. This paper performed a comprehensive analysis of features to enhance stroke prediction effectiveness. A reliable dataset for stroke prediction is taken from Kaggle website to gauge the effectiveness of the proposed algorithm. The dataset has a class imbalance problem which means a total number of negative samples is higher than a total number of positive samples. The results are reported based on a balanced dataset created using oversampling techniques. The proposed work used Smote and Adasyn to handle imbalanced problem for better evaluation metrics. Additionally, the hybrid Neural Network and Random Forest (NN-RF) utilizing the balanced dataset by using Adasyn oversampling achieves the highest F1-score of 75% compared to the original unbalanced dataset and other benchmarking algorithms.