<![CDATA[Diabetes mellitus (DM) is a global health problem that affects the quality of life and life expectancy of patients. Diabetes risk prediction can help in disease prevention and management. This study aims to optimize ensemble classification for diabetes risk prediction using a Genetic Algorithm (Optimize Selection). The Ensemble methods used are Bagging, Random Forest, and AdaBoost. The genetic algorithm is applied for Ensemble model hyperparameter optimization. The data used is the Pima Indians Diabetes dataset which consists of 768 samples with 8 features. Experimental results show that Ensemble Classification optimized with the Genetic Algorithm (Optimize Selection) produces quite good performance. The accuracy of the Genetic Algorithm (Optimize Selection) + Ensemble Bagging Classification Algorithm got a result of 97.14%, the Genetic Algorithm (Optimize Selection) + Random Forest Ensemble Classification Algorithm got a result of 98.57%, and the Genetic Algorithm (Optimize Selection) + AdaBoost Ensemble Classification Algorithm got a result of 99.87%. %, These results indicate that this approach can be an effective solution in diabetes risk prediction.]]>
Copyrights © 2024
