Web search engines compete to offer the fastest responses with highest relevance. However, as Web collections grow, it becomes more difficult to achieve this purpose. As most user...
Sheffield's participation in the inaugural Arabic cross language track is described here. Our goal was to examine how well one could achieve retrieval of Arabic text with the...
In this paper, we describe ideas and related experiments of Tsinghua University IR group in TREC 2004 QA track. In this track, our system consists three components: Question analy...
In this paper, a system for Named Entity Recognition in the Open domain (NERO) is described. It is concerned with recognition of various types of entity, types that will be approp...
We report on the results of a series of experiments with a prototype text retrieval system which uses relatively advanced natural language processing techniques in order to enhanc...