Abstract:
The annual statistics of the incidence of diabetes mellitus shows a stable growth. Most diabetics are not aware of their diagnosis and the risks that this disease carries. Also, an important problem is playing sports in diabetes, it worries both people with newly diagnosed diabetes and those who live with this disease for a long time. Thus, early diagnosis and treatment are critical to prevent morbidity. The main goal of the study is to assess the risk of diabetes among people depending on their lifestyle and marital status. This study proposes several data mining models for predicting type 2 diabetes mellitus (T2DM). Diabetes prediction models are very accurate with rates above 95%, which is why their application is very relevant to healthcare applications. Two sets of data were used to conduct machine learning, one is the main one, the second was used for validation.