Colloquium Details

Online Data Forensics for Click Fraud Detection

Speaker: Ahmed Metwally, Google

Location: Warren Weaver Hall Room 102

Date: February 16, 2010, 12:30 p.m.

Host: Zvi Kedem

Synopsis:

Internet advertising is crucial for the thriving of the entire Internet, since it allows producers to advertise their products, and hence contributes to the well being of e-commerce. In addition, the advertising revenue stream supports the operation of search engines and Internet content publishers. Some publishers are dishonest, and use automation to generate traffic to defraud the advertisers. Similarly, some advertisers automate clicks on the advertisements of their competitors to deplete their competitors' advertising budgets. These fraudulent clicks jeopardize the entire Internet advertising model.

In this talk, we describe the Internet advertising model, and discuss the issue of click fraud that is an integral problem in such a setting. We start by classifying the click fraud techniques into two major classes based on the motivation of the fraudulent publishers and advertisers. We describe traffic analysis problems that model detecting both classes of fraud attacks. We propose using streaming and sampling algorithms on aggregate traffic as a viable way of detecting automated traffic, while not violating the surfers’ privacy. We conclude by reporting the results of deploying these algorithms on various networks of advertising commissioners.

Speaker Bio:

Dr. Ahmed Metwally received his M.S. and Ph.D. degrees in computer science from the University of California at Santa Barbara, in 2006 and 2007, respectively and his B.S. degree in computer and systems engineering from Alexandria University in 2000.

Dr. Metwally’s research focuses on databases, and the new applications of data management and data mining. He has publications in the areas of data streams, data privacy, mining association rules, OLAP and data cubes and the implementation of data management algorithms on specialized hardware. His Ph.D. dissertation proposed data management techniques for analyzing Internet advertising traffic as a comprehensive approach for advertisement click fraud detection.

Throughout his PhD, Dr. Metwally worked at several Internet Advertising Commissioners including Google, Ask.com, Valueclick, Fastclick, and Commission Junction. Since April 2008, he has been with the Ad Traffic Quality team at Google.

Notes:

Refreshments will be offered starting 15 minutes prior to the scheduled start of the talk.


How to Subscribe