Subscribe Now Subscribe Today
 
FOLLOW US:     Facebook     Twitter
   
Journal of Applied Sciences
  Year: 2011 | Volume: 11 | Issue: 18 | Page No.: 3301-3307
DOI: 10.3923/jas.2011.3301.3307
An Online Model on Evolving Phishing E-mail Detection and Classification Method
Ammar Ali Deeb Al-Momani, Tat-Chee Wan, Karim Al-Saedi, Altyeb Altaher, Sureswaran Ramadass, Ahmad Manasrah, Loai Bani Melhiml and Mohammad Anbar

Abstract:
Phishing e-mails pose a serious threat to electronic commerce as they are used broadly to defraud both individuals and financial organizations on the Internet. Criminals lure online users into revealing their passwords or account numbers by sending e-mails as if they come legitimately from financial organizations; these users in turn update and/or provide their account and billing information. In the current study, propose a novel concept that adapts the Evolving Clustering Method for Classification (ECMC) to build new model called the Phishing Evolving Clustering Method (PECM). PECM functions are based on the level of similarity between two groups of features of phishing e-mails. PECM model proved highly effective in terms of classifying e-mails into phishing e-mails or ham e-mails in online mode, speed and use of a one-pass algorithm. PECM also proved its capability to classify e-mail by decreasing the level of false positive and false negative rates while increasing the level of accuracy to 99.7%. The model was built to work in online mode and can learn continuously without consuming too much memory because it works on a one-pass algorithm. Therefore, data are accessed one time from the memory and then the rule is created depending on the evolution of the profile if the characteristics of the phishing e-mail have been changed. PECM is a clustering-based learning model that adaptive Evolving Clustering Method to distinguish between phishing e-mails and ham e-mails in online mode.
 [Fulltext PDF]   [Fulltext HTML]   [XML: Abstract + References]   [References]   [Report Citation]
 RELATED ARTICLES:
  •    Fast-Flux Botnet Detection Based on Weighted SVM
  •    A SVM-based Technique to Detect Phishing URLs
  •    A Comparative Study of Clustering Algorithms
  •    An Agent-based Model for Agriculture E-commerce System
  •    Information Systems for Enhancing Customer Relationships
  •    Machine Learning Technique Approaches in Drug Discovery, Design and Development
How to cite this article:

Ammar Ali Deeb Al-Momani, Tat-Chee Wan, Karim Al-Saedi, Altyeb Altaher, Sureswaran Ramadass, Ahmad Manasrah, Loai Bani Melhiml and Mohammad Anbar, 2011. An Online Model on Evolving Phishing E-mail Detection and Classification Method. Journal of Applied Sciences, 11: 3301-3307.

DOI: 10.3923/jas.2011.3301.3307

URL: http://scialert.net/abstract/?doi=jas.2011.3301.3307

 
COMMENT ON THIS PAPER
.