Comparison of Machine Learning Algorithms in Diabetes Risk Classification

Zairy Cindy Dwinnie (Universitas Islam Negeri Sultan Syarif Kasim Riau, Indonesia)
Zaira Cindya Dwynne (Universitas Islam Negeri Sultan Syarif Kasim Riau, Indonesia)
Mohammed Jahidul Islam (Niels Brock Copenhagen Business College, Denmark)
Noviarni Noviarni (Universitas Islam Negeri Sultan Syarif Kasim Riau, Indonesia)

Indonesian Journal of Applied Technology and Innovation Science
Vol. 1 No. 2 (2024): IJATIS August 2024

Diabetes is a disease in which blood sugar levels are excessive without insulin control so that body functions do not function normally. Diabetes is also a disease that many people suffer from and is one of the main causes of death throughout the world. For this reason, we need to know the factors that are indicators of someone suffering from diabetes. This research compares the Decision Tree, Logistic Regression, and K-Nearest Neighbors algorithms with accuracy and Confusion Matrix parameters to determine diabetes sufferers in 520 data with the main indicator attributes supporting diabetes. From the test results of the three algorithms, the Decision Tree and K-Nearest Neighbors models have the highest accuracy of 86%. The Logistic Regression Algorithm has a fairly good accuracy of 83%.