Analysis of Attribute Selection and Classification Algorithm Applied to Hepatitis Patients

Abstract

Data mining techniques are widely used in classification, attribute selection and prediction in the field of bioinformatics because it helps to discover meaningful new correlations, patterns and trends by sifting through large volume of data, using pattern recognition technologies as well as statistical and mathematical techniques. Hepatitis is one of the most important health problem in the world. Many studies have been performed in the diagnosis of hepatitis disease but medical diagnosis is quite difficult and visual task which is mostly done by doctors. Therefore, this research is conducted to analyse the attribute selection and classification algorithm that applied to hepatitis patients. In order to achieve goals, WEKA tool is used to conduct the experiment with different attribute selector and classification algorithm . Hepatitis dataset that are used is taken from UC Irvine repository. This research deals with various attribute selector namely CfsSubsetEval, WrapperSubsetEval, GainRatioSubsetEval and CorrelationAttributeEval. The classification algorithm that used in this research are NaiveBayesUpdatable, SMO, KStar, RandomTree and SimpleLogistic. The results of the classification model are time and accuracy. Finally, it concludes that the best attribute selector is CfsSubsetEval while the best classifier is given to SMO because SMO performance is better than other classification techniques for hepatitis patients.