Workshop on Link Analysis,
Counterterrorism and Security
Submission deadline: PASSED
Notification of acceptance: PASSED
Camera ready copy due: PASSED
Proceedings (SIAM will make an archival copy of the proceedings available in a few weeks)
Final submissions should meet the same formatting requirements as the main conference, namely: Papers should have a maximum length of 12 pages (single-spaced, 2 column, 10 point font, and at least 1" margin on each side). Authors should use US Letter (8.5. x 11.) paper size.
However, initial submissions can be in any reasonable format, although reviewers may stop reading when they feel length is becoming excessive. Submissions in pdf form should be emailed to : email@example.com. The use of an author's name as part of the file name would be appreciated.
Under development, but includes:
This workshop is the third workshop on this topic at the SIAM International Data Mining Conference. Information about the 2004 workshop, including proceedings, can be found here. The workshop attracts a mixture of academics, and security and counterterrorism practitioners.
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 may also wish to participate, either directly in the workshop or in an industrial track in the main conference.
This year a special focus will be the Enron email dataset, made public by the U.S. Department of Justice. This is the largest real email corpus in the public domain, and makes an ideal target for link analysis, social network analysis, and textual analysis. This dataset resembles the kind of data collected as part of counterterrorism efforts: some participants wished to conceal both the extent of their connections to each other and the content of their discussions; while the vast majority of the messages were completely innocent. Hence techniques that are effective on this dataset might be expected to perform well in counterterrorism settings also. We anticipate that about half of the presentations will be related to this dataset.
Although this dataset has been made public, most of the email authors were not involved in any wrongdoing. We ask participants to be sensitive to the feelings of these authors whose privacy has been impacted by the actions of others.
The dataset is available here (zipped and tarred, about 350 MB). The dataset is very large: about half a million emails among several thousand individuals.
Other processed versions of the dataset are available here. Please point out errors in these versions to us, and make your own versions available if possible.
The workshop is an integral part of the SIAM International Data Mining Conference, which takes place from 21-23rd April 2005, and no separate registration is required.