Prediction causes of diabetes detection using machine learning methods

Diabetes is one of the most common chronic diseases in the world, affecting about 530 million people. The main causes of its occurrence include genetic predisposition, obesity, im-proper eating behavior, insulin resistance and bad habits. Early detection of the disease can prevent its development. Many symptoms of diabetes, such as dry mouth, frequent urination, blurred vision, weight loss, constant hunger, are not always immediately considered as signs of the disease. But these symptoms can be early indicators of high blood glucose levels. The paper analyzes the factors and causes that affect the risk of developing diabetes and makes predictions using the Decision Tree, Random Forest, k-NN and Ada Boost machine learning methods. The results are analyzed and the accuracy of the methods used is assessed. The re-sults obtained will allow for the detection of significantly more cases of diabetes before it oc-curs, early and effective treatment, and reduction of healthcare costs.