Diabetes Mellitus depicts a group of chronic metabolic disorders affecting more than 450 million people worldwide. Diabetes is a hazardous and incurable disease. The root cause of this health complication is yet to be ascertained. Many researchers have worked on this area to make it easier for medical professionals to examine it. In this study, recent literature was analyzed. It is to be accounted that different Machine Learning models have obtained better results and performs well in many diabetes-associated tasks. This research investigates the causes of readmission and hospital readmission within 30 days of deliverance amongst diabetic patients. The proposed work is implemented on the Diabetes Readmission dataset acquired from the UCI Repository in the area of medical science contains 55 attribute and 100000 records of the patients. The model obtains benchmarked score which is 98% for the stacking model and Long Short-Term Memory (LSTM) gives 98.5% precision. The framework assists in classifying whether the diabetic patient will be readmitted to the hospital within 30 days or not.