LAADM 2010 : Workshop on Link Analysis for Adversarial Data Mining
Call For Papers
April 29 -- May 1, 2010
The Columbus, A Renaissance Hotel
at the SIAM International Conference on Data Mining.
Submission deadline: January 2
Notification of acceptance: February 1
Camera ready copy due to Antonio Badia: February 14
School of Computing
skill at cs.queensu.ca
Department of Computer Engineering and Computer Science
University of Louisville
abadia at louisville.edu
2009 Program Committee:
The Program Committee is still being formed; accepted so far:
Jafar Adibi, PricewaterhouseCoopers.
Daniel Barbarà, George Mason University.
Hans Chalupsky, ISI.
John Gersh, John Hopkins University.
Lise Getoor, University of Maryland.
Mark Goldberg, RPI.
Mark Maybury, MITRE.
Antonio Sanfilippo, PNNL.
Simeon Simoff, University of Technology, Sydney.
Bülent Yener, RPI.
Daniel Zeng, University of Arizona.
Submit by email to abadia at louisville dot edu please. It would be helpful if you could name your attached file in a way that includes the name of one of the authors. Both full papers (6-9 pages) and short papers (3-4 pages) will be considered.
See the Submissions page of the main conference web page for details of format (do not submit to the main conference site).
Besides academia, we encourage submissions from research centers, industry, and government organizations.
This workshop is the sixth workshop on this topic at the SIAM International Data Mining Conference. The workshop name has changed from "Workshop on Link Analysis, Counterterrorism and Security" to emphasize a widening of the topic. The workshop attracts a mixture of academics; security, law-enforcement and counterterrorism practitioners, and data analysts from several industries.
An Adversarial setting is one where the subject of analysis (the "adversary") is actively engaged in trying to remove, disguise or otherwise alter traces of activity. Thus, the data given is not a passive collection of facts, but it may have been actively manipulated. Furthermore, an intelligent adversary may change behavior over time, in an attempt to further confuse the analyst. This poses new challenges to all data mining activities, but in particular to link analysis, since in this area the correctness and completeness of data play an important role. Traditional link analysis algorithms are not developed to cope with the challenges of an adversarial setting. Note that such setting is frequent in real-life applications of data mining; they include (cyber)security, law enforcement, counterterrorism, fraud detection, phising, spam filtering, junk email detection, and threat detection. Thus, advances in adversarial link analysis is a relatively new area that has wide applicability.
The workshop provides a venue in which to present early work in relevant areas. An online proceedings will be created, and hardcopy proceedings will be available to conference attendees (and through SIAM afterwards). However, authors may retain copyright.
Industrial organizations with products that are suitable for analyzing large datasets in an Adversarial setting may also wish to participate, either directly in the workshop or in an industrial track in the main conference.
Topics of Interest:
Relational data mining
Social Network Analysis
Dynamic network analysis
Web mining applied to (cyber)security, law enforcement, counterterrorism, fraud detection, phising, spam filtering, junk email detection, threat detection
Text mining applied to security, law enforcement, counterterrorism, counterterrorism, fraud detection, phishing, spam filtering, junk email detection, threat detection
Visualization of link structures
Performance evaluation measures for Adversarial settings
Innovative applications related to link analysis
Papers on any of these issues, and related topics, are welcome provided that their primary focus is one of the topics of interest.
This will be a full-day workshop. It will consist of an invited talk, a series of presentation from accepted papers, and a panel discussion. The panel will be organized around the workshop topics and guided by the content of the presented papers.
The workshop is part of the SIAM International Data Mining Conference, which takes place from April 29-May 1, 2010 in Columbus, Ohio. At least one author from each accepted paper is expected to register and attend the workshop to present their work.