Prediction of Diabetes Using Data Mining Technique: A Review

Abstract

Data mining plays an efficient role in prediction of diseases in health care industry. Diabetes has become one of the major global health problems at present. According to the WHO 2014 report, around 422 million people worldwide are suffering from diabetes. Diabetes is a metabolic disease where the improper management of blood glucose levels led to risk of many diseases like heart attack, kidney disease, eye etc. Many algorithms have been developed for the prediction of diabetes and accuracy estimation. This paper gives detailed evaluation of existing data mining methods used for prediction of diabetes. More than twenty research papers had been referred during this research work. And according to those research papers, decision tree related algorithms had been used in most of the research works and this algorithm has given the best performance also. So that, to have the best accuracy and performance in data mining related projects, decision tree algorithm and related algorithms can be used. Further, it gives some idea about the tools which can be used in data mining. WEKA, Orange, MATLAB, Tanagra and Rapid Miner are some of the data mining tools which are commonly used. Some of the researchers have used more than one tool for their research works. But almost all of the referred papers have used the data mining tool WEKA. Finally, this paper gives the idea that, using Decision tree related algorithms may give high performance and high accuracy in data mining related works.