SRV: A Striking Model based on Meta-Classifier for Improving Diagnosis Type 2 Diabetes

Diagnosis of diabetes is a classification problem that attracts more in recent years. Diabetes mellitus happens when the whole body cannot provide an adequate quantity of insulin to adjust glucose levels. In the low insulin level, food products in glucose are turned into glucose, improving the sugar to a more than average level. All existing works show that many techniques are successful for this disease, Artificial Intelligence. There exist many classification models to aim the prediction of diabetes. We introduce a novel model to investigate the role of pre-processing and data reduction for classification problems in the diagnosis of diabetes. The model has four steps consisting of Pre-processing, Feature sub-selection, Classification, and Performance. In the classification technique, we apply the voting technique with three classifiers. Many experiments were conducted to reveal the performance of the proposed work for the diagnosis of diabetics. The results confirmed the superiority of our model over its counterparts, and the best accuracy, precision, recall, and F1 were achieved at 96.67, 100, 100, and 94.01%, respectively.