Donna Harman and Ellen Voorhees, of the Information Access Division (IAD), participated in the recent Cross-Language Evaluation Forum (CLEF 2001) held in Darmstadt, Germany September 3-4, 2001. CLEF, a European-funded activity, is a continuation and expansion of the cross-language evaluation work started in 1997 at ITL's Text REtrieval Conference (TREC), focusing on the European languages. NIST continues as one of four partners in this evaluation, providing both the software for all languages and the data collection activity needed in all tasks requiring use of the English language. Harman and Voorhees serve on the Workshop Steering Committee. More information may be found at the CLEF Web site at http://galileo.iei.pi.cnr.it/DELOS/CLEF.

Following the TREC model, the CLEF series of system evaluation aims at promoting research and development in Cross-Language Information Retrieval (CLIR) by providing an infrastructure for the testing and evaluation of information retrieval systems operating on European languages in both monolingual and cross-language contexts, and creating test-suites of reusable data which can be employed by system developers for benchmarking purposes. At this Workshop, results of the CLEF activity were discussed. Researchers and developers compared performance between systems using different cross-language strategies. These tasks included the retrieval of documents written in one of four different European languages (German, French, Italian, and Spanish) using questions written in English, and conversely, the retrieval of English documents from questions written in different European languages.

More information about TREC may be found at http://trec.nist.gov.