Author : Jitendra Sheetlani, Ajay Vyas, Harsh Gupta [ Volume No.:XII, Issue No.IV-July 2023] [Page No : 25-29] [2023]

Diabetes has become a significant global public health concern as the prevalence of non-communicable diseases continues to rise. Heart disease claims the lives of approximately 18 million people annually, with diabetes and high blood pressure emerging as primary contributing factors. The social, physical, and economic consequences of diabetes are substantial. Elevated blood sugar levels characterise this chronic condition and occur due to the body’s inability to produce or properly respond to insulin. Data mining enables analysts to efficiently analyse extensive data sets to identify patterns and trends associated with diabetes. In recent years, machine learning (ML) methods have been utilised for diabetes prediction. Data mining involves extracting essential information and leveraging it to enhance dynamic effectiveness. Various AI techniques, including Support Vector Machine (SVM), Random Forest, Decision Tree (DT), K-Nearest Neighbor (KNN), Artificial Neural Network (ANN), and Naive Bayes (NB) classifiers, have been employed in diabetes prediction. This research focuses on analysing different machine-learning models for diabetes prediction. The paper is structured into sections: the first section discusses the various machine learning models and their distinct activation functions employed in this study. In contrast, the second section presents a comparative analysis of these models.