Type 2 Diabetes Prediction using Gray Wolf Optimization Algorithm

Abstract

Background: Increasing the number of diabetic patients and the ignorance of most of these patients about the dangers arising from it is a challenge that threatens human lives. Materials and Method:In this paper, a new solution based on the Gray Wolf Optimization (GWO) algorithm for predicting type 2 diabetes is presented. The main purpose of the proposed method is to increase the accuracy of prediction and also to reduce the probability of getting stuck in local optimal points. In more detail, the proposed method consists of two parts: 1-data preprocessing including data preparation and noise cancellation and 2-data classification using gray wolf algorithm. The Pima Indians Diabetes dataset in MATLAB simulation environment was used to analyze the data and compare the research results. Results:The simulation results show that by adjusting the parameters of the gray wolf algorithm, about 6% better prediction accuracy is obtained than other researches. Conclusion:Also, for a more accurate evaluation of the proposed method, two other datasets have been used for testing. The results of experiments show that the proposed model for health management in diabetes is effective.