An efficient feature selection method for classification in health care systems using machine learning techniques

Abstract

Data mining can be used for a large amount of applications. Among one is the health care systems. Usually, medical databases have large quantities of data about patients and their medical history. Analyzing this voluminous data manually is impossible. But this medical data contain very useful and valuable information which may save many lives if analyzed and utilized properly. Data mining technology is very effective for Health Care applications for identifying patterns and deriving useful information from these databases. Diabetes is one of the major causes of premature illness and death worldwide. In developing countries, less than half of people with diabetes are diagnosed. Without timely diagnoses and adequate treatment, complications and morbidity from diabetes rise exponentially. India has the world's largest diabetes population, followed by China with 43.2 million. This paper describes about the application of data mining techniques for the detection of diabetes in PIMA Indian Diabetes Dataset (PIDD). In this paper we propose a Feature Selection approach using a combination of Ranker Search method. The classification accuracy of 81% resulted from our approach proves to be higher when compared with previous results