Authors: S. Carolin Jeeva; Elijah Blessing Rajsingh
Addresses: Department of Computer Applications, Karunya University, Coimbatore, India ' Department of Computer Sciences Technology, Karunya University, Coimbatore, India
Abstract: Phishing is an online scandalous act that occurs when a malevolent web page impersonates as legitimate web page in the intension of exploiting the confidential information from the user. Phishing attack continues to pose serious risk for web users and annoying threat in the field of electronic commerce. Feature selection is the process of removing unrelated features and thus reduces the dimensionality of the feature. This paper focuses on identifying the foremost features that categorise legitimate websites from phishing websites based on feature selection. In real world identifying phishing URL with low computational time and accuracy is very important and thus feature selection is considered in this work. A comparative study is carried out on different data mining classifiers before and after feature selection and the performance are evaluated in terms of accuracy and computational rate. The results indicate that the proposed approach detects phishing websites with considerable accuracy.
Keywords: web security; cyber-crime; phishing; Attribute selection; classification and machine learning.
International Journal of Electronic Security and Digital Forensics, 2017 Vol.9 No.2, pp.116 - 131
Received: 02 Mar 2016
Accepted: 31 Oct 2016
Published online: 07 Apr 2017 *