Comparison Algorithm for Diabetes Classification with Consideration of Mutual Information and Information Feature

Diabetes is a prevalent disease in humans that caused by excessive sugar levels the body. If left untreated, it can lead to severe consequences such as paralysis, decay certain parts of body, and even death. Unfortunately, early detection diabetes difficult, many cases go untreated until too late. However, development technology has opened up new possibilities for treatment diabetes. One approach classification, commonly used method field Computer Science. Classification various fields, including health, agriculture, animal diseases, draw conclusions based on input data using cause-and-effect relationships. Many different learning concepts methods be with Decision Tree concept being one most popular examples. This study compares several classification methods, Tree, Random Forest, AdaBoost, Stochastic Gradient Boost, feature selections carried out MI IF. The aims evaluate effectiveness these influence selection improving their performance. Based results study, concluded Mutual Information Importance Feature improve accuracy some particularly Boost. algorithm did not show any improvement after selection. best was achieved Boost original dataset without selection, while Forest showed highest all features. Overall, suggest useful technique performance algorithms prediction. suggests future research could investigate other Neural Network or Deep Learning, use optimization like Genetic Algorithm Particle Swarm Optimization results.