TDT 2002 is the fifth in a series of open evaluations that have investigated several aspects of developing algorithms for the automatic organization of news stories by the real-world events that they describe. Researchers who are interested in the following topics are encouraged to find out more about TDT 2002: text filtering, cross-language issues, machine translation, speech recognition, text segmentation, compensating for degraded quality text, novelty detection, or other similar topics.

TDT evaluation tasks for 2002 are currently expected to be

• New event detection (Formerly First Story Detection)
• Story link detection
• Segmentation
• Topic detection
• Topic Tracking

Other tasks are being considered and the final evaluation specifications are being debated.

The preliminary call for participation is available in either MS-word or postscript.

EVALUATION PROJECT STATUS

There is an evaluation planned for 2002. The participants will receive the data September 2, 2002 and have a month to process the data. See the TDT schedule for details.

To prepare for the evaluation, NIST will be administering a Dry Run Evaluation during June and July of 2002. The Dry Run Evaluation is designed to help new participants prepare their evaluation systems early in the process and so make sure they fully understand the evaluation methodology.

Data for the evaluation are provided by the Linguistic Data Consortium (LDC). It can be either purchased as a member of the LDC, or provided free for the evaluation with some restrictions. See the TDT resources web page for additional information.

Joining the TDT mailing list will keep you abreast of recent information.

CONTACT INFORMATION

Questions regarding the TDT corpora and obtaining access to it should be directed to ldc@ldc.upenn.edu