A Data Mining Approach for Genetic Diabetes Prediction

Abstract

Diabetes is one of the major cause of death in recent decades which occur at any age. Diabetes is a disease that occurs when your blood glucose, also called blood sugar, is out of range. There are many reasons for occurring diabetes like lifestyle problem, other diseases, medicine, pregnancy, genetic problem etc. We have worked here with genetic diabetes and done a data mining approach for predicting diabetes. Data mining tools proves successful result in case of diseases diagnosis. There are different data mining techniques available like Tracking patterns, Classification, Association, Outlier detection, Clustering, Regression, Prediction. We work here with prediction technique to make a data mining approach for the diabetes patient which occurred genetically. This prediction is done across different ranges of age men and women who have diabetes. In the dataset if his/her parents or grandparents have diabetes, he/she has treated as genetic diabetes patient. We also do the prediction of gender wise diabetes patients who take insulin. Then also predict the insulin taking diabetes patients for different ranges of age man and woman. This research will open a new platform to research. The limitation of the research is where diabetes create for other disease or other unusual activities, there the proposed system can’t be applicable. But the research can be used for giving more awareness on genetic diabetes and giving a new prediction which can be used in medical field.