Hybrid Prediction Model for Type-2 Diabetes Mellitus using Machine Learning Approach

Diabetes is a disease that affects millions of people throughout the world. Diabetes is a chronic illness with no treatment available. As a result, early detection is essential. In this paper author uses Machine Learning Approach (MLA) such as Decision Tree (DT), Random Forest (RF) and Logistic Regression (LR) to predict Type2 Diabetes Mellitus (T2DM). The recent advancements in technology and continuous progress have changed the landscape of healthcare with changes in lifestyle and rise in living standards. Diabetes remains the leading cause of death globally for early prediction of T2DM. Artificial Intelligence (AI) tools are used for early T2DM and prognosis. However, it is still in its nascent stage when it comes to the early prediction of Diabetes. Author Proposed a Methodology Framework for Type2 Diabetes prediction and other health conditions. The first part of this paper aims at developing and implementing a prediction model based on various Diabetic stages. Using predictive analysis on the dataset author applied three ML algorithms to predict T2DM. Author finds that a model combining Logistic Regression (LR), Decision Tree (DT) and Random Forest (RF) is effective at predicting Diabetes. The result shows the Logistic Regression algorithm has the highest accuracy of 99.349% as compared to DT, RF respectively. In order to improve the classification accuracy in research work will help practitioners in the early detection of T2DM.