Diabetes Classification Using Machine Learning Techniques

Abstract

Machine learning techniques play an increasingly prominent role in medical diagnosis. With the use of these techniques, patients’ data can be analyzed to find patterns or facts that are difficult to explain, making diagnoses more reliable and convenient. The purpose of this research was to compare the efficiency of diabetic classification models using four machine learning techniques: decision trees, random forests, support vector machines, and K-nearest neighbors. In addition, new diabetic classification models are proposed that incorporate hyperparameter tuning and the addition of some interaction terms into the models. These models were evaluated based on accuracy, precision, recall, and the F1-score. The results of this study show that the proposed models with interaction terms have better classification performance than those without interaction terms for all four machine learning techniques. Among the proposed models with interaction terms, random forest classifiers had the best performance, with 97.5% accuracy, 97.4% precision, 96.6% recall, and a 97% F1-score. The findings from this study can be further developed into a program that can effectively screen potential diabetes patients.