Uncovering Risk Factors for Diabetes Mellitus Using Data Mining Classification Techniques

Diabetes Mellitus (DM), a growing global health concern, demands innovative strategies for early detection and management. This study employs advanced data mining classification techniques, such as Decision Trees, Random Forest, Support Vector Machines, and Neural Networks, to analyze a comprehensive dataset of patient demographics, medical histories, and laboratory test results. Our aim was to identify significant predictors for DM and enhance predictive models for its early diagnosis. Through rigorous evaluation metrics like accuracy, precision, recall, and F1 score, we assessed the performance of these algorithms, supplemented by mathematical expressions that clarify the classification mechanisms. Results revealed that these data mining techniques effectively distinguish between diabetic and non-diabetic individuals, with the Random Forest algorithm showing superior performance in balancing precision and recall, particularly suited to the imbalanced nature of medical datasets. This study not only underscores the potential of data mining in healthcare but also offers critical insights into the comparative advantages of each classification technique. Such insights guide healthcare professionals in selecting the most appropriate models for early DM detection, contributing to the advancement of DM care and prevention strategies. This research paves the way for future innovations in personalized and preventive healthcare through computational techniques.