Ghana Graduates Conference

Origin-Based Classification of Advance Fee fraud Electronic Mails Using e-STAT

Purpose The fact that typical e-mailers will always include their true e-mail addresses to facilitate replies can assist in classifying electronic mails.

Design/methodology/approach We checked the IP-addresses from which spam mails originate as  one way of ascertaining their actual origin. This provides the needed information that can aid in building a database of spam mails to be blacklisted so as to stop the delivery of further mails from such addresses in future.

Findings From the corpus of e-mails tracked using E-Stat, we obtained a blacklist of about 2,173 e-mail sender’s addresses, 563 URLs within spam mails and a total of 13,491 bag-of-words common to Advance Fee Fraud spam e-mails

Research implications This research developed a domain specific content analysis tool (e-STAT) using Bayesian technique.

Practical and Policy implications The analysis of mail origin using e-STAT revealed some elements of current concept drift among Advance Fee Fraudsters

Three Learning Points for Ghana and Africa Specific Spam corpus from Ghana and other African Countries can be treated with the same methodology in order to develop learning algorithms that can track spam mails

Keywords: Bayesian, E-mails, IP Address, Spam.

Abayomi-Alli Adebayo lectures at the Dept of Electrical and Computer Engineering, Igbinedion University Okada, Nigeria. His research is centered around cyber security. He can be contacted at This e-mail address is being protected from spambots. You need JavaScript enabled to view it