Published Paper Details:

Diabetes, a major global source of morbidity, must be identified early in order to be effectively managed and complications avoided. In this work, we propose a machine learning method for diabetes prediction that makes use of a Support Vector Machine (SVM) model. We carried out comprehensive preparation using a publically accessible dataset, which included choosing pertinent properties, addressing missing values, and normalizing features. Several kernels (linear, polynomial, and RBF) were used to optimize the SVM model, and performance was enhanced through hyperparameter adjustment. The model's effectiveness was validated using evaluation measures such as accuracy, precision, recall, F1-score, and ROC-AUC, which achieved [certain performance criteria]. Our results demonstrate the potential of SVM in clinical decision support systems for the diagnosis of diabetes. In order to improve model resilience, future research will examine sophisticated feature engineering approaches and broaden the dataset to include a variety of populations.