Nature Inspired Algorithms Along With Support Vector Machine for Detecting Web Phishing Attack
No Thumbnail Available
Date
2018-06-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Addis Ababa University
Abstract
The advancement of Computer Technology has grasped attention of individuals living on the 21th century. This advancement has made life much simpler where one can use phones in his/her hand to access different kinds of facilities without even making a few yards of movement.
Web Technology is one example that has materialized to this generation due to the invention of the Internet. The Technology allows Internet users to access different kinds of services in a modern and easy way. A user only needs a web browser software or an application that allows him/her to perform bank transaction without being physically present on the bank compound.
Even though efforts made to enhance the technology, there are still tradeoffs on keeping user’s personal information private and in a secured manner. This has been a major problem since the innovation of the Web.
There are different kinds of security threats on Web Technology mainly; Web based Threats, Denial of Service Attack, and Phishing.
Phishing is a web based attack that works by creating a fake version of the real sites web interface to gain the users trust. It is a combination of social engineering and technical exploits designed to convince a victim to provide personal information, usually for monetary gain of the attacker. Phishing has become the most popular practice among the crimes performed on the web especially on e-commerce services and social media.
Besides, the attack is becoming more frequent and sophisticated which makes it difficult to tackle. This is causing a significant impact on both service provides and consumers since it involves the risk of identity theft and financial losses that leads to loss of trust by service customers.
In this thesis we have proposed a novel classification approach which uses Support Vector Machine (SVM) optimized with Firefly Algorithm (FA) to detect phishing attack. The parameters of SVM are optimized with the help of FA for detecting phishing attack by a specific website. We implemented a prototype web browser which can be used as an agent for the purpose of retrieving Web URL and reporting phishing attack to user. We have found an acceptable range of accuracy and minimized error rate on the classification. The user testing also shows a convenient way of reporting the attack.
Description
Keywords
Web, Website, Phishing, Phishy, Phishtank, Nature Inspired Algorithm, Firefly Algorithm, Support Vector Machine