The Power Of Data Analysis - "Spamalytics"

Some of you may have seen the article in the New York Times by John Markoff (endnote1) announcing a paper to be presented at last week’s IEEE conference. This paper is an update to research conducted by a team at the International Computer Science Institute in Berkeley, California. The institute is associated with the University of California, San Diego and the University of California, Berkeley. A paper published by the team in 2008 Spamalytics: An Empirical Analysis of Spam Marketing Conversion outlines interesting research in the area the research team has coined as “spamalytics.”

The paper describes a methodology to understand the architecture of a spam campaign and how a spam message converts into a financial transaction. The team looks at the “conversion rate” or the probability an unsolicited email will create a sale. The team uses a parasitic infiltration of an existing botnet infrastructure to analyze two spam campaigns: one designed to propagate a malware Trojan, the other marketing online pharmaceuticals. The team looked at nearly a half billion spam emails to identify:

  • the number of spam emails successfully delivered
  • the number of spam emails successfully delivered through popular anti-spam filters
  • the number of spam emails that elicit user visits to the advertised sites
  • the number of “sales” and “infections” produced
Read more