An Ensemble Model of Data Mining Approach for Enhancing the Performance of Diabetes Mellitus Diagnosis

Abstract

Data mining is a significant creditworthy technique in fields such as banking, communication, education, advertising and healthcare. Since data mining is a remarkable resource in the domain of medical databases, our research focuses on employing ensemble data mining techniques on medical databases.Of many chronic diseases, diabetes mellitus is slowly but surely becoming a major threat to all ages across the globe and a central public health issue. In order to prevent and control the diabetes, several researches are ongoing. The main intend of this paper is to introduce a new ensemble approach, which plays a key role in the domain of  medical data mining, more than ever, in detecting diabetes mellitus. Ensemble methodology is one of the growing tactics for strengthening classifier accuracy. Typically, ensemble is an effective technology that combines multiple foundation classifier predictions. Support vector machine, Random Forest and Adaboost are very efficient algorithm and play a vital role in delivering the highest level of precision rate. When in ensemble of these algorithms offer more accuracy than when applied in single. Here an ensemble technique has been applied to improve the accuracy rate of diagnosing diabetes mellitus.To boost the accuracy rate of diagnosis of diabetes mellitus, an integrated strategy was adopted here in this paper.