Diabetic Disease Prediction System using Associative Classification with Improved Classifier Accuracy

Abstract

 Associative Classification (AC) is a novel and rewarding strategy in data mining that incorporates association mining and classification to construct classification models. AC is power of constructing efficient and accurate classification systems that are easy to interpret by end-user. AC makes use of more simple technique to predict the health problems like diabetics. Generally the data in medical field is enormous and unstructured related to various diseases in human being. It is very difficult to analyze and anticipate the disease problems that change their severity time to time. This method of classification combines with association rule mining helps to anticipate vast amount of data and also construct the efficient classification model. In this research paper, we shed the light on the need and use of AC on vast and unstructured data in order to predict diabetic disease. The association rules obtained in training phase of the data is used to predict the diabetic disease using predictive Apriori.CBA which is one of associative classification strategy is conveyed for accuracy measures.