S. Annas, A. Aswi, M. Abdy, B. Poerwanto
This study aims to determine the factors that significantly affect the classification of stroke. The response variable used is the type of stroke, namely non-hemorrhagic stroke and hemorrhagic stroke. The predictors used were cholesterol level, blood sugar level, temperature, length of stay, pulse rate, and gender. By using logistic regression, the results obtained modeling accuracy of 74.8% where the predictors that have a significant effect (alpha <0.05) are cholesterol level and length of stay. © 2021 Institute of Physics Publishing. All rights reserved.
Statistics Department, Universitas Negeri Makassar, Indonesia; Mathematics Department, Universitas Negeri Makassar, Indonesia
