TREC 2009 Million Query (1MQ) Track

The Million Query track is running for the third time in 2009. Its goal is to study evaluation and optimization of information retrieval systems over very many very incompletely judged topics organized into predefined categories.

Mailing list:	trec-million09@udel.edu
	Subscribe at https://udel.edu/mailman/listinfo/trec-million09
Coordinators:	Ben Carterette (University of Delaware)
	Virgil Pavlu (Northeastern University)
	Hui Fang (University of Delaware)

Features of the 2009 track include:

• A large new web corpus - ClueWeb09
• An optional new task - predicting the type of information need (precision-oriented or recall-oriented) the query represents and serving it accordingly
• Prioritized queries so that sites can participate by running as few as 1,000 queries or as many as 40,000 queries
• An experimental design for studying reusability of small sets of judgments

Latest News