Title:
Application of Machine Learning Techniques for Effective Diabetes Management

Authors:
Seun Mayowa Sunday, United Kingdom

Abstract:

The study employs machine learning techniques to predict and manage type 2 diabetes (T2DM). Various machine learning models, including Decision Tree, Support Vector Machine (SVM), Artificial Neural Network (ANN), and Gradient Boosting Machine (GBM), were assessed. The research involved extensive data preparation, feature selection using mutual information, and hyper parameter tuning via Grid Search CV. The GBM and SVM models demonstrated superior performance, achieving high accuracy and AUC values. Feature importance analysis identified glucose, BMI, and diabetes pedigree function as critical predictors. This study highlights the potential of machine learning to enhance diabetes management and advocates for future research to explore diverse datasets and advanced algorithms. The practical implications suggest significant improvements in personalized treatment plans and early intervention strategies for T2DM patients.

DOI: http://dx.doi.org/10.51505/ijaemr.2025.1517