Parallel Information Extraction System for Multilingual Information Access

Information Extraction1 is mostly a monolingual technology. This feature is evident when considering the most important worldwide initiative focusing on Information Extraction, i.e. the Message Understanding Conference. In this competition English has always been the unique target language, with the exception of MUC-6 (MET-1) where Spanish and Chinese were considered as well. Even the introduction of these languages, however, has not significantly changed the monolingual status of the competition, as all the systems were assumed to be completely independent. This situation contrasts with Information Retrieval where, almost since the beginning ([1]) multilinguality was one of the important research tracks, as attested by the introduction of a Cross-Language IR track in TREC-6.