Diabetes is a serious disease in which your body cannot properly control the amount of sugar in your blood because it does not have enough insulin. Diabetes is the most common medical complication during pregnancy, representing 3.3% of all live births. In this, we have a dataset of approximately 1000 people. The decision tree is obtained from Python using which we can predict whether the people present in the dataset suffer from diabetes or not. Diabetes is a disease in which your blood glucose, or blood sugar, levels are too high. When you are pregnant, high blood sugar levels are not good for the baby. Classification of the probability of diabetes is done based on various factors. The main aim of this work is the detection of Diabetes Mellitus using different models and classifies the data as diabetic and non-diabetic. Our health care systems are rich in information but they are poor in knowledge so there is a large need of having techniques and tools for extracting the information from the huge data set so that medical diagnosis can be done. Data Mining is a process of semi-automatically analyzing large databases to find useful patterns. Data mining attempts to discover rules and patterns from data as it deals with large volumes of data, stored primarily on disk. Data mining mainly deals with knowledge discovery in databases.