Decisions and Machine Learning for Diabetes Diagnosis System

The Medical Diagnosis System of Diabetes aiming to identify the correct diagnosis of Patient’s diabetes as quickly as possible and at lower cost. The Diabetes Diagnosis System (DDS) has three subsequent stages; the first stage is the construction of medical dataset (MD) with eight features taken for 1000 patients and covering three classes (Diabetic, Non-Diabetic, and Predicted- Diabetic). The second stage is the data mining- based machine learning, which introduces Interactive Dichotomizer 3 (ID3) classifier; is employed to diagnose the condition of a patient from his/her medical history. The outcome of implementing the proposal system showed that the accuracy of the ID3 model has been found approximately (98.25).