Analyzing Diabetes Risk Factors Using Support Vector Machine Algorithm

Abstract

Diabetes mellitus is a persistent metabolic condition that presents considerable health issues globally.  Active prediction and risk assessment are essential for mitigating problems and enhancing patient outcomes.  This research examines diabetes risk variables with the Support Vector Machine (SVM) algorithm, an effective supervised machine learning method appropriate for classification applications.  Diverse clinical and lifestyle characteristics, such as age, body mass index (BMI), glucose levels, blood pressure, and familial history, are used as input features.  The SVM model is developed and evaluated to categorize people as diabetic or non-diabetic with high precision.  Performance assessment is conducted with criteria accuracy, recall, and F1-score.  The results indicate that SVM proficiently detects critical risk variables and delivers dependable predictions, underscoring its potential as a decision-support instrument in healthcare for the early identification of diabetes.