Accepted Papers

Automating Financial Surveillance.

Milosavljevic M, Delort JY, Hachey B, Arunasalam B, Radford W, Curran J. (Submission 14.)

Monitoring Social Attitudes using Rectitude Gains.

Wawer A. (Submission 12.)

A proposal for a multilingual epidemic surveillance system.

Lejeune G, Hatmi M, Doucet A, Huttunen S, Lucas N. (Submission 3.)

Cross-Lingual Analysis of Concerns and Reports on Crimes in Blogs.

Nakasaki H, Abe Y, Utsuro T, Kawada Y, Fukuhara T, Kando N, Yoshioka M, Nakagawa H, Kiyota Y. (Submission 9.)

Automated Event Extraction in the Domain of Border Security.

Atkinson M, Piskorski N, Tanev H, van der Goot E, Yangarber R, Zavarella V. (Submission 15.)

Security Level Classification of Confidential Documents written in Turkish.

Alparslan E., Bahsi H. (Submission 4.)

Signaling Events in Text Streams.

Schuhmacher J., Koster C. (Submission 2.)

The Workshop will be a full-day event, on December 9th.
(Program to be announced soon, approximately 09:00-17:00)

Workshop Theme

The vast and growing amount of user-generated textual content, including online news streams, blogs, electronic encyclopedias (e.g., the Wikipedia), and other openly accessible and dynamically changing data readily available on the Web has led to the emergence of new approaches to extracting valuable, structured, and previously unknown information from such data. The aim of this workshop is to bring together researchers from academia and industry who develop technologies for mining open-source user-generated textual data on the Web, as well as end-users interested in exploiting such technologies for knowledge discovery. The emphasis is placed on large-scale text mining systems and application-oriented approaches to processing on-line textual content in the context of security-related applications. Examples of such applications include:

• global medical and epidemic surveillance,
• conflict early warning,
• early detection of man-made or environmental hazards,
• risk assessment,
• border surveillance,
• cross-border crime detection,
• terrorism,
• other applications relevant for public health, security, and law enforcement institutions

Due to a multitude of challenges of diverse and complex nature that are related to automating the process of mining user-generated content on the Web, we believe that this workshop will serve as a forum to bring together researchers from different areas, including data mining, language technology, computational linguistics, information sciences, information retrieval and Web mining, for sharing ideas and discussion. In particular, we believe that there is an important gap to be filled, since the aforementioned research communities have had limited interaction previously in the context of the topic of the Workshop. The second major goal is to engage governmental and inter-state user communities, and to bring them together with scientists and funding agencies.

Topics of interest

• Mining from news streams, blogs, document repositories, and other openly accessible and dynamically changing data, including Web 2.0 content, for the purpose of identifying threats to security or public health,
• Emphasis on multilingual approaches, and work on languages other than English,
• Applications, such as information extraction, classification, summarization, sentiment detection, event detection, event forecasting, trend detection, information fusion, and more,
• Contributions in the form of applications (working systems and prototypes) as well as theoretical results are welcome,
• Application domains include crisis-related event reporting, political and environmental analysis, and medical intelligence, under the general umbrella of the security intelligence domain,
• Methods including machine learning, rule-based, and hybrid approaches.

NB: Please note, this Workshop welcomes all work on computational approaches to the analysis of textual data for gathering information from openly accessible sources only. Submissions that focus on legal questions stemming from snooping, spying, privacy infringement or violation, etc., will not be considered relevant to the Theme of the Workshop, and the Committee will not be able to review them.

Workshop Format

We invite papers addressing primarily the language technology, natural language processing, data mining and information retrieval communities, as well as the relevant end-user groups. Submissions are invited in two categories:

• (a) regular: 6 pages long, research papers presenting novel approaches and solutions, and
• (b) short (posters): 4 pages long, system demonstrations, descriptions, and work in progress.

The workshop will feature one prominent expert as an invited speaker from academia/industry.

It is our intention to initiate a cyclic event on this topic. We believe that such an event will be met with interest by the community due to a key combination of factors: the ever-growing need for applications in the security domain and the wide availability of open-source textual data on the Web. The workshop aims at bringing together technology enablers and end users, which would benefit the User-Centric Media community.