Portal de Periódicos da CAPES

An Extensive Approach Towards Heart Stroke Prediction Using Machine Learning with Ensemble Classifier

2021; Springer Nature; Linguagem: Inglês

10.1007/978-981-16-5747-4_66

2524-7573

Divya Paikaray, Ashok Mehta,

Retinal Imaging and Analysis

Heart stroke remains one of the eminent diseases which has a great impact on the mortality rate. Besides the other diseases which may be diagnosed and treated, heart stroke is mostly a quick occurring event with minimal time for response. With the advancement of technology in the medical field, diagnosis and analysis are getting accurate. With proper analysis of certain attributes, stroke can be predicted well in time before the occurrence, and life may be saved. One such attempt has been made in this paper using machine learning algorithms. A dataset was acquired from Kaggle. The attributes provided in the dataset were analyzed using nine algorithms of machine learning, namely; linear discriminant analytics, Logistic regression, Gaussian Naive Baye's, Support vector machine, K-nearest Neighbor's classifier, Random forest classifier, bagging classifier, Ada boost classifier, and Gradient Boosting classifier. The pattern of the attributes as per the provided dataset was monitored for accurate prediction of heart stroke in the patients. The experimental data were divided into training and testing datasets for further analysis and comparison. Post analysis of the result, it was established that the Random Forest algorithm was trustworthy with 95.10% of accuracy.