Data - English Test Questions (Topics)

TREC calls a natural language statement of information need a "topic" to distinguish it from a "query", which is the data structure actually presented to the retrieval system. The topics are formatted using a very simple SGML-style tagging. Different topic sets have different fields included in the topic statements. In addition to the elimination or addition of fields, the format of the topic statements may vary slightly from file to file.

The topics have been divided into files based on the TREC task they were used in. Relevance judgements for these topics against various portions of the TREC collections are available at Data-English Relevance Judgements. The documents are covered by intellectual property agreements and are for use by TREC participants only (see Data-English Documents page).