This paper describes the Unisys MUC-3 text understanding system, a system based upon a three tiered approach to text processing in which a powerful knowledge-based form of information retrieval plays a central role . This knowledge-based form of information retrieval makes it possible to define a n effective level of text analysis that falls somewhere between what is possible with standard keyw...

Multilingual Information Retrieval usually forces a choice between free text indexing or indexing by means of multilingual thesaurus. However, since they share the same objectives, synergy between both approaches is possible. This paper shows a retrieval framework that make use of terminological information in free-text indexing. The Automatic Terminology Extraction task, which is used for thes...

This paper describes the SpamBayes submissions made to the Spam Track of the 2005 Text Retrieval Conference (TREC). SpamBayes is briefly introduced, but the paper focuses more on how the submissions differ from the standard installation. Unlike in the majority of earlier publications evaluating the effectiveness of SpamBayes, the fundamental ‘unsure’ range is discussed, and the method of removi...

Video-Text pre-training aims at learning transferable representations from large-scale video-text pairs via aligning the semantics between visual and textual information. State-of-the-art approaches extract features raw pixels in an end-to-end fashion. However, these methods operate frame-level directly thus overlook spatio-temporal structure of objects video, which yet has a strong synergy wit...

In this paper we present an information retrieval system that indexes only site anchor text to verify the efficiency of reference information in a navigation retrieval task. We propose two relevancy measures to maximize limited information: reference consistency and specificity of word combination. Our results show that navigation retrieval with a site anchor text can pinpoint highly relevant d...

In cross-language text retrieval, query text objects in one language are matched against a collection of text objects in another. Previous work showed that a two-language cross-language text-retrieval system can be created completely automatically by a method called CL-LSI trained on an aligned corpus of text objects. In this paper, we look at the challenge of creating a three-language cross-la...