Abstract
Stroke is one of the main causes of long-term disability and death around the world. For its significant impact, stroke is also defined as a medical emergency, where immediate treatment is vital to save patients’ lives. For these reasons, an efficient prediction plays an important role in stroke prevention and cure. In this paper, attempting to implement a model predicting stroke efficiently, logistic regression and random forest algorithms are adopted, as well as a stroke dataset. They are trained and make predictions with the preprocessed dataset independently. Multiple evaluation indicators are employed to evaluate the two models’ results. Comparisons between their performances and the reasons for their discrepancies are both introduced, based on which the more suitable one is chosen as the final model. Models’ bias and variance and how they influence the results are discussed as well. In addition, some helpful propositions and approaches to improve the model’s performance will also be introduced.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call