Diabetes Diagnosis Using Machine Learning Algorithms: Methods and Techniques

Diabetes is a sickness with no clear solution, thus early detection is essential. During our study, we employed data mining, machine learning techniques to forecast diabetes. The data is on 768 participants and their different characteristics. On the dataset, we applied four machine learning algorithms for the prediction of diabetic condition. The implemented algorithms were examined for the robustness and consistency and compared using accuracy and F-1 score methods. It was found that the Support Vector Machine prototype works well for prediction of diabetic condition with an accuracy of 79% accuracy. This research strives for suggesting a suitable method to help the doctors and health professionals for early detection of diabetes.