Multilingual Topic Detection and Tracking: Successful Research Enabled by Corpora and Evaluation

By Charles L. Wayne. 2000

156 citations recorded by the source corpus.

Topic Detection and Tracking (TDT) refers to automatic techniques for locating topically related material in streams of data such as newswire and broadcast news. DARPA-sponsored research has made enormous progress during the past three years, and the tasks have been made progressively more difficult and realistic. Well-designed corpora and objective performance evaluations have enabled this success.

Subfields: Artificial Intelligence, Statistical and Nonlinear Physics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Jimmy J. Lin
2. Daniel Marcu
3. B. Taskar
4. Jennifer Gillenwater
5. Alex Kulesza
6. Christopher D. Manning
7. Wen-tau Yih
8. Mark Stevenson

In Statistical and Nonlinear Physics

1. Alex Kulesza
2. Jimmy J. Lin
3. B. Taskar
4. Daniel Marcu
5. Miles Efron

Citations in the corpus

• Determinantal Point Processes for Machine Learning (2012, methodology): “1 Related Work Research from to the Topic Detection and Tracking (TDT) program [154] has led to useful methods for tasks like link detection, topic detection, and topic tracking that can be seen as subroutines for graph threading on text collections.”
• Assessor Differences and User Preferences in Tweet Timeline Generation (2015, background): “At a high level, the tweet timeline generation task bears a family resemblance to topic detection and tracking (TDT) [33, 3], “other” nuggets in question answering [32, 9, 8], nugget-based evaluations in DARPA’s BOLT program, 1 and temporal summarization at TREC [12, 5].”
• Building a Discourse-Tagged Corpus in the Framework of Rhetorical Structure Theory (2001, background): “Annotation ranges from broad characterization of document-level information, such as topic or relevance judgments (Voorhees and Harman, 1999; Wayne, 2000) to discrete analysis of a wide range of linguistic phenomena.”
• Discovering Diverse and Salient Threads in Document Collections (2012, background): “The Topic Detection and Tracking (TDT) program (Wayne, 2000) has recently led to some research in this direction.”
• The Reuters Corpus Volume 1 -from Yesterday’s News to Tomorrow’s Language Resources (2002, background): “By contrast, RCV1 covers a complete year of editorial output, allowing the investigation of temporal issues such as topic detection & tracking (Wayne, 2000), conflict prediction (Bond et al, 1997) or financial market forecasting (Giles et. al.”