PENERAPAN MACHINE LEARNING UNTUK PREDIKSI PENYAKIT STROKE

Abstract

Stroke is a global health problem and one of the leading causes of adult disability. Early detection and prompt treatment are needed to minimize further damage to the affected brain area and complications to other parts of the body. Machine learning techniques can be used to predict stroke detection. Machine learning algorithms such as Naive Bayes, Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Decision Tree are compared in this study to obtain the best performance in predicting stroke. The implementation stages in this research consist of the pre-processing data, the application of the algorithm and the evaluation and analysis. The Naïve Bayes algorithm obtains better Accuracy, Precision, Recall, and F1-Measure values compared to other algorithms. The values of Accuracy, Precision, Recall, and F1-Measure obtained by Naïve Bayes are 93.93%, 88.23%, 93.93%, and 91.00%, respectively. So the conclusion of this study is that the Naïve Bayes algorithm has the best performance compared to the SVM, KNN and Decision Tree algorithms in predicting stroke.
Keywords: decision tree, klasifikasi, k-nearest neighbor, naïve bayes, stroke, support vector machine