International Journal of the Computer, the Internet and Management Vol.22 No.1 (January-April, 2014) pp 62-67 62 Decision Tree Classifier for Computer Self-Efficacy with Best First Feature Selection Wanphen Wirojcharoenwong Department of Information Technology, Faculty of Information Technology, King Mongkut's University of Technology North Bangkok, Wongsawang, Bangsue, Bangkok wanphen.wi@gmail.com Natthaphong Luangnaruedom Department of Computer Technology, Faculty of Technology, Siam Technology College 46 Jarunsanitwong 10 rd., Thapra branch, Bangkok-yai district, Bangkok. 10600 demontee_smileface@hotmail.co.th Montean Rattanasiriwongwut Department of Information Technology Management, Faculty of Information Technology, King Mongkut's University of Technology North Bangkok, Wongsawang, Bangsue, Bangkok montean@it.kmutnb.ac.th Monchai Tiantong Department of Computer Education, Faculty of Technical Education, King Mongkut's University of Technology North Bangkok, Wongsawang, Bangsue, Bangkok monchai@kmutnb.ac.th Abstract - This research proposes CfsSubsetEval subset evaluator and using Best First feature selection algorithm to select appropriate features and classification of data mining techniques by Decision Tree (C 4.5). We adjust the confident factor value to 5 values: 0.1, 0.25, 0.5, 0.6 and 1.0 to test for the confident factor from any value to the highest accuracy. We used dataset from questionnaire data which was collected from 1,894 undergraduate students and 17 attributes. The results showed the attributes from feature selection has 12 attributes and that the confident factor value 0.6 is high accuracy and confident factor value 0.1 is low accuracy. Keywords - Computer Self-Efficacy, Feature selection, Decision Trees I. INTRODUCTION In 1977, Bandure was the first writer to use the term self-efficacy. The concept of self-efficacy is most widely used measure the computer self-efficacy are presented [4]. Studies of computer self-efficacy has a major impact on an individual’s expectations toward using computer [5]. In recent years, researchers have tried to bring out the statistics that describe the level of computer self-efficacy. In this paper, it proposed data mining techniques for describe the level of computer self-efficacy. The article proceeds as follows: In the next section, we review the background information about computer self-efficacy. In section 3, we present the related work. In section 4, an overview of feature selection and C4.5 Decision Tree techniques. In section 5, we present the experimental framework and results and reference in section 6. II. BACKGROUND Self-Efficacy and Computer Self-Efficacy Self-efficacy is defined as the personal judgment about one’s capability to adopt certain behaviors and actions in order to