The Use of Bagging Technique to Optimize PSO Based C4.5 to Diagnose Diabetes

Diabetes is one of the diseases with the highest number of sufferers worldwide. In addition, diabetes is in the top 10 leading causes of death and has been significantly increasing since 2000. If not treated quickly, diabetes will be very dangerous for the human organs, especially the parts of the body traversed by blood vessels. Therefore, this study aims to detect early stage diabetes. So that the treatment of diabetes is faster before it develops into a more serious direction. The begging technique on the PSO-based C4.5 algorithm is applied to determine the accuracy of this algorithm model in classifying people with diabetes. The bagging method and Particle Swarm Optimization (PSO) aim to address the weaknesses of the C4.5 algorithm in overlapping data, a large number of attributes, and classes. According to the experiments, the bagging technique on the PSO-based C4.5 algorithm produces a better accuracy value by approximately 95.58%, while the C4.5 algorithm ware 90.38%.