The Topic Detection and Tracking (TDT) 2001 evaluation project is the fourth annual open evaluation of TDT Technologies. As an evaluation community within the DARPA Translingual Information Detection and Summarization (TIDES) Program, TDT's task is to advance the state of the art in technologies that organize events in simultaneously collected streams of multilingual news data. The evaluations are open to all interested participants.

The 2001 TDT evaluation collmenated in a two-day workshop sponsored by NIST. The presentations and system description papers from the workshop have been collected on this site. At the workshop, plans were made for the next evaluation TDT 2002.

BACKGROUND

A call for participation was posted in March to the TDT mailing list. A dry run evaluation was ran in the Summer of 2001. First time participants were required to complete a dry run.

The formal evaluation will be held in the Fall of 2001. See the TDT schedule for details. Data for the evaluation are provided by the Linguistic Data Constorium (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.

Results of last year's evaluation were reported at the TDT 2000 workshop held in Gaithersburg, Maryland, immediately after the TREC evaluation workshop. Some findings from the workshop were reported out at the HLT 2001 Conference, in March 2001.

CONTACT INFORMATION

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