Machine Vision Inspection Systems, Image Processing, Concepts, Methodologies, and Applications

10A Review on Phishing—Machine Vision and Learning Approaches

Hemamalini Siranjeevi¹, Swaminathan Venkatraman²* and Kannan Krithivasan³

¹ Department of Computer Science and Engineering, Srinivasa Ramanujan Centre, SASTRA Deemed University, Kumbakonam, India

² Discrete Mathematics Laboratory, Department of Mathematics, Srinivasa Ramanujan Centre, SASTRA Deemed University, Kumbakonam, India

³ Department of Mathematics, SASTRA Deemed University, Thanjavur, India

Abstract

Phishing is a criminal act wherein fraudsters acquire users’ sensitive information. Criminals create a lot of phishing websites to lure users into traps that seize their credentials. Many countermeasures have evolved, but they fail because of high false prediction rates. This review presents a detailed report on some of the attempts toward avoiding, detecting, and preventing phishing. The literature has documented the role of training and education on the reduction of phishing victims. It also shows that out of all the methods used for identifying phishing websites, visual methods demand high computational power.

Keywords: Phishing, identity-theft, fraud-detection, fake websites, illegitimate user

10.1 Introduction

Online transactions have encountered an enormous loss of about £21.6 million in the first half of 2012, which is an increase of 28% compared with 2011. India’s loss amounts to $225 million in 2013, says RSA’s Fraud report 2013. About 46,747 phishing attacks created a loss of nearly $453 million during ...