Predicting Diabetes Using Supervised Machine Learning Algorithms

Abstract This study explores the application of supervised machine learning algorithms in predicting diabetes, a chronic and life-threatening disease with no definitive cure. We employed three supervised machine learning algorithms which include logistic regression, Random Forest, and k-Nearest Neighbors (KNN) to develop predictive models and identify significant features contributing to diabetes. These models were evaluated based on accuracy, minimal test error, specificity, and sensitivity. Our analysis reveals that KNN achieved the highest performance with accuracy of 88.36%, sensitivity of 87.88%, and specificity of 88.83% among the algorithms tested, making it the most reliable for diabetes prediction in this dataset. The goal of this research is to aid in early detection and management of diabetes by leveraging machine learning on electronic health records.