Decision Support System for Diabetes Classification Using Data Mining Techniques: Classification Using Data Mining Techniques

Decision Support System for Diabetes Classification Using Data Mining Techniques: Classification Using Data Mining Techniques

Ahmad M. Al-Khasawneh (Hashemite University, Jordan)
DOI: 10.4018/978-1-5225-5460-8.ch012

Abstract

The use of data mining algorithms in health information systems has played a significant role in developing applications that help to diagnose different diseases. The type of the disease determines the selection of the algorithm, parameters to be used, and dataset pre-processing steps, etc. In this chapter, diagnosing diabetes mellitus is the target since it has gained significant attention in the last few decades due to the increased severity of the disease. Four predictive data mining approaches are being used in diagnosing diabetes. Four models were implemented to diagnose diabetes from PIMA dataset: k-nearest neighbor, support vector machine, multilayer perceptron neural network, and naive Bayesian network. Giving the highest classification accuracy, support vector machine technique outperformed the others with a value of 78.83%.
Chapter Preview
Top

Background

Many researchers applied data mining techniques in solving clinical problems, predictive models are utilized to help in disease diagnosis process, specifically, the diabetes illness. A survey was conducted by (Tomar and Agarwal, 2013) on data mining approaches (classification, regression, and clustering approaches) used in healthcare. Classification models predict the class of a new observation among predefined categories of the target variable, while in regression modelling the output is a numeric value (Williams, 2011).

Predictive data mining builds both classification and regression modelling using several algorithms such as decision trees, random forests, boosting, support vector machines, linear regression, neural networks, naive Bayesian classifier, Bayesian networks, and the k-nearest neighbours' models (Bellazzi & Zupanb, 2008; Williams, 2011; Al-Khasawneh & Hijazi 2014). Pradhan et al. (2011) introduced general guidelines to design a predictive model for diagnosing diabetes using hybrid of soft computing and data mining techniques

Complete Chapter List

Search this Book:
Reset