 This paper presents a novel approach for improving the accuracy of diabetes classification using a metaheuristic optimization algorithm. The proposed approach uses a dynamic albiruni earth radius and dipper-throated optimization algorithm, BIRDTO, to select relevant features from the dataset and then classifies them using a random forest classifier. The results show that the proposed approach outperforms existing approaches in terms of accuracy and statistical significance. This article was authored by Amel Ali El-Husson, Abdel Aziz A. Abdel Hamid, S. K. Tofek and others.