ASSESSMENT OF FORECASTING SYSTEMS FOR PATIENT IDENTIFICATION: EXPLORING MACHINE LEARNING APPROACHES FOR PREMATURE DIABETES RECOGNITION

Undiagnosed diabetes can worsen its already significant effects on the body, resulting in complications such as diabetic nephropathy and strokes. On a global scale, a significant number of individuals face challenges related to this disease, highlighting the essential requirement for timely identification in order to preserve one's well-being. The exponential rise in global diabetes cases renders this a substantial worry. Machine Learning (ML) has the potential to provide accurate projections and profound insights through using expertise and analyzing data.Current research is utilizing machine learning (ML) to investigate diabetes within the Pima Indian culture. Scientists are use the data manipulation software R to detect and analyze trends and patterns in risk variables. Their objective is to categorize patients into diabetes and non-diabetic groups by creating and evaluating five distinct predictive models using R. Supervised machine learning techniques, including Radial Basis Function (RBF) kernel support vector machines, Artificial Neural Networks (ANN), k-nearest neighbors (KNN), Multifactor Dimensionality Reduction (MDR), and linear and RBF kernel support vector machines, are utilized in this process.