AIRWeb '05

First International Workshop on
Adversarial Information Retrieval on the Web

At the 14th International World Wide Web Conference (WWW2005), 10-14 May 2005, Chiba, Japan

Workshop sponsored by

Welcome

Dear Participant:

Welcome to the First International Workshop on Adversarial Information Retrieval on the Web (AIRWeb). This workshop is intended to bring together researchers and practitioners that are concerned with the on-going efforts in adversarial information retrieval on the Web. We have a total of eight peer-reviewed papers to be presented -- five research presentations and three synopses of work in progress. All convey the latest results in adversarial web IR, and address topics such as web spam, blog spam, cloaking, redirection, link optimization for PageRank, automated link spam detection, link bombs, reverse engineering of ranking algorithms, and propaganda. In addition, we will have a panel session in which workshop participants may raise additional questions of interest to industry experts and researchers.

I extend my thanks to the authors and presenters, and to the members of the program committee for their work in contributing to the material that forms an outstanding first workshop. I also sincerely thank Ask Jeeves for their support of the workshop, enabling us to partially cover travel costs for many of the student presenters. It is my hope that you will find the work presented interesting enough that you will ask questions, contribute ideas, and perhaps get involved in future work in this area.

Brian D. Davison, Program Chair
Lehigh University, Bethlehem, PA
19 April 2005

Overview

Search is the single most common application used on the Web. The attraction of hundreds of millions of searches per day provides significant incentive to content providers to do whatever necessary to rank highly in search engine results. The use of techniques that push rankings higher than they belong is often called spamming a search engine (or spamdexing). Such methods typically include textual as well as link-based techniques. Like e-mail spam, search engine spam is a form of adversarial information retrieval; the conflicting goals of accurate results of search providers and high positioning by content providers provides an interesting and real-world environment to study techniques in optimization, obfuscation, and reverse engineering, in addition to the application of information retrieval and classification.

The AIRWeb'05 workshop solicited technical papers on any aspect of adversarial information retrieval on the Web. Particular areas of interest included, but were not limited to:

search engine spam and optimization,
crawling the web without detection,
link-bombing,
reverse engineering of ranking algorithms,
advertisement blocking, and
web content filtering.

Papers addressing higher-level concerns (e.g., whether 'open' algorithms can succeed in an adversarial environment, whether permanent solutions are possible, etc.) were also welcome.

Authors were invited to submit papers and synopses in PDF format. We encouraged submissions presenting novel ideas and work in progress, as well as more mature work. Submissions were judged by multiple experts on relevance, significance, originality, clarity, and technical merit.

Program

9:00am	Welcome
9:15am	Session 1 Web Spam Taxonomy Zoltán Gyöngyi, Stanford University Hector Garcia-Molina, Stanford University Synopsis: An Analysis of Factors Used in Search Engine Ranking Albert Bifet, Technical University of Catalonia Carlos Castillo, University of Chile Paul-Alexandru Chirita, L3S Research Center Ingmar Weber, Max-Planck-Institute for Computer Science Synopsis: Web Spam, Propaganda and Trust Panagiotis T. Metaxas, Wellesley College Joseph DeStefano, College of the Holy Cross
10:30am	Morning Break
11:00am	Session 2 Pagerank Increase under Different Collusion Topologies Ricardo Baeza-Yates, ICREA-Univ. Pompeu Fabra and University of Chile Carlos Castillo, Universitat Pompeu Fabra Vicente López, Universitat Pompeu Fabra Cloaking and Redirection: A Preliminary Study Baoning Wu, Lehigh University Brian D. Davison, Lehigh University Synopsis: Optimal Link Bombs are Uncoordinated Sibel Adali, Renssalaer Polytechnic Institute Tina Liu, Renssalaer Polytechnic Institute Malik Magdon-Ismail, Renssalaer Polytechnic Institute
12:30pm	Lunch
2:00pm	Expert Panel Session Moderator: Brian D. Davison, Lehigh University Panelists: Andrei Broder, IBM Research Soumen Chakrabarti, IIT Bombay David Cohn, Google Tim Converse, Yahoo Marc Najork, Microsoft Research
3:30pm	Afternoon Break
4:00pm	Session 3 Blocking Blog Spam with Language Model Disagreement Gilad Mishne, University of Amsterdam David Carmel, IBM Research Ronny Lempel, IBM Research SpamRank -- Fully Automatic Link Spam Detection András A. Benczúr, Hungarian Academy of Sciences (MTA SZTAKI) and Eötvös University Károly Csalogány, Hungarian Academy of Sciences (MTA SZTAKI) and Eötvös University Tamás Sarlós, Hungarian Academy of Sciences (MTA SZTAKI) and Eötvös University Máté Uher, Hungarian Academy of Sciences (MTA SZTAKI)
5:00pm	Discussion Directions for future work Organization of AIRWeb '06

Proceedings

Full Papers

Blocking Blog Spam with Language Model Disagreement, by Gilad Mishne, David Carmel and Ronny Lempel
Cloaking and Redirection: A Preliminary Study, by Baoning Wu and Brian D. Davison
Pagerank Increase under Different Collusion Topologies, by Ricardo Baeza-Yates, Carlos Castillo and Vicente López
SpamRank -- Fully Automatic Link Spam Detection, by András A. Benczúr, Károly Csalogány, Tamás Sarlós, and Máté Uher
Web Spam Taxonomy, by Zoltán Gyöngyi and Hector Garcia-Molina

Synopses

An Analysis of Factors Used in Search Engine Ranking, by Albert Bifet, Carlos Castillo, Paul-Alexandru Chirita and Ingmar Weber
Optimal Link Bombs are Uncoordinated, by Sibel Adali, Tina Liu and Malik Magdon-Ismail
Web Spam, Propaganda and Trust, by Panagiotis T. Metaxas and Joseph DeStefano

Program Committee

Program Chair

Brian D. Davison, Lehigh University

Program Committee Members

Andrei Broder, IBM Research
David Carmel, IBM Research Haifa
Tim Converse, Yahoo
Nick Craswell, Microsoft Research Cambridge
Matt Cutts, Google
Dennis Fetterly, Microsoft Research
David Gibson, IBM Research
David Hawking, CSIRO
David D. Lewis, Independent Consultant
Mark Manasse, Microsoft Research
Kevin McCurley, IBM Research
Urban Mueller, Search.ch
Marc Najork, Microsoft Research
Jan Pedersen, Yahoo
Bernhard Seefeld, Search.ch
Baoning Wu, Lehigh University
Tao Yang, Ask Jeeves/Univ. of California, Santa Barbara

Last modified: 3 May 2005, Brian D. Davison