Prediction of diabetes using data mining and machine learning

Diabetes is a chronic metabolic complication in which the lack of proper regulation of bloodglucose levels in diabetic patients leads to the risk of heart attack, disease and kidney failure. Andbesides, it causes serious health problems such as fatal kidney damage or blindness. It may leadthe patient to death. There is still no exact treatment for this disease, but it can be controlled withmedication and diet. In this way, the importance of correct diagnosis of diabetes is very importantto identify diseases in the early stages and take precautionary measures. A lot of data has beenaccumulated on this topic because there are so many patients with this condition. It provides thepossibility for researchers to use data mining techniques in this topic. The advancement in thefield of computer provides a large amount of data. The main role of data analysis is to input andobtain the required data that can be used with various data mining techniques. Diagnosing diabetesis an important and difficult role in medicine. By combining computer knowledge and medicalscience, more accurate and faster information can be obtained to predict and treat diabetes. In thisarticle, we will examine the data, algorithms and results, and at the end, the sample work doneby me will be explained. It should be noted that the results obtained in this article were obtainedby WEKA ۳.۸.۶ software and ۱۰ algorithms (Navie Bayse, Jrip, J۴۸, Random Forest, Bayes Net,Bagging, IBK, SMO, OneR, K*) Has been studied . In this research, we also use India’s PIMAdata as a source and this research was conducted in ۲۰۲۱ and was implemented at Qom Universityof Technology.