Medical decision support system based on artificial immune recognition immune system (AIRS), fuzzy weighted pre-processing and feature selection Kemal Polat * , Salih Gu ¨nes ß Selcuk University, Engineering Faculty, Department of Electrical and Electronics Engineering, 42075 Konya, Turkey Abstract In this study, diagnosis of hepatitis disease, which is a very common and important disease, was conducted with a machine learning system. The proposed machine learning approach has three stages. The first stage, the feature number of hepatitis disease dataset was reduced to 10 from 19 in the feature selection (FS) sub-program by means of C 4.5 decision tree algorithm. Then, hepatitis disease dataset is normalized in the range of [0, 1] and is weighted with fuzzy weighted pre-processing. Then, weighted input values obtained from fuzzy weighted pre-processing is classified by using AIRS classifier system. In this study, fuzzy weighted pre-processing, which can improved by ours, is a new method and firstly, it is applied to hepatitis disease dataset. We took the dataset used in our study from the UCI machine learning database. The obtained classification accuracy of our system was 94.12% and it was very promising with regard to the other classification applications in the literature for this problem. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Artificial immune recognition immune system; Fuzzy weighted pre-processing; Feature selection; Hepatitis disease; Medical diagnosis 1. Introduction Most of the time the hepatitis diagnoses is made by a rou- tine blood testing or during a blood donation. The hepatitis is a viral infection that also was transmitted by blood or blood products in the past, when there was no test available to screen for this infection. Risk factors are as follows: blood transfusions, tatoos and piercing, drug abuse, hemodyalisis, health workers, sexual contact with hepatitis carrier (http:// www.angelfire.com/biz2/physician29/hepatitis.html, last arrived: 20 January 2006). The use of classifier systems in medical diagnosis is increasing gradually. There is no doubt that evaluation of data taken from patient and decisions of experts are the most important factors in diagnosis. But, expert systems and different artificial intelligence techniques for classifica- tion also help experts in a great deal. Classification systems, helping possible errors that can be done because of fatigued or inexperienced expert to be minimized, provide medical data to be examined in shorter time and more detailed. In this study a new medical diagnosis method was pro- posed to be used hepatitis disease diagnosis problem as a classifier. This method involves a three-stage system in which an artificial immune recognition system, fuzzy weighted pre-processing and feature selection are hybrid- ized. The first stage of the whole system is used to feature selection for feature reduction in hepatitis diseases. In the second stage, fuzzy weighted pre-processing that firstly proposed by ours is used to weighting the whole dataset. And then in the third stage, artificial immune recognition system as classification method is used to medical decision making for hepatitis disease. The used data source is taken from the University of California at Irvine (UCI) Machine Learning Repository (UCI Machine Learning Repository). This dataset is com- monly used among researchers who use machine learning 0957-4174/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.eswa.2006.05.013 * Corresponding author. Tel.: +90 332 2232098; fax: +90 332 2410635. E-mail addresses: kpolat@selcuk.edu.tr (K. Polat), sgunes@selcuk. edu.tr (S. Gu ¨nes ß). www.elsevier.com/locate/eswa Expert Systems with Applications 33 (2007) 484–490 Expert Systems with Applications