The World Health Organization classified diabetes as a major global health problem and a leading cause of death worldwide. Studies have recently shown an increase in the epidemiology of diabetes owing to the increase in the related complications to diabetes, such as cardiac attacks and deaths. The prediction of medical datasets at the early stage is normally done using machine learning techniques (MLT) in a bid to save life. Different data repositories host different medical datasets, which can be accessed for real-world application. Machine learning (ML) can provide the necessary techniques as its major mission is disease prediction from datasets. This thesis presents a machine learning model which Artificial Neural Network (ANNs) trained using enhanced grey wolf optimizer (EGWO), for the classification of diabetes on publicly available "Pima Indian Diabetes (PID) dataset". Several experiments have been executed on this dataset with variation in size of population, techniques to handle missing data and their impact on classification accuracy have been discussed. Finally, the results are compared with other nature inspired algorithm trained ANN. EGWO attained better results in terms of classification accuracy than the other algorithms. In addition, it was better than the original version of GWO. The convergence curve proved that EGWO has balanced the local and the global search abilities, because it was faster to reach better positions than the original GWO. In conclusion, the proposed algorithm trained the ANN and enhanced the classification accuracy. It can be used for predicting different medical case studies, or different machine learning tasks.