An Association Thesaurus for Information Retrieval

Although commonly used in both commercial and experimental information retrieval systems, thesauri have not demonstrated consistent bene ts for retrieval performance, and it is di cult to construct a thesaurus automatically for large text databases. In this paper, an approach, called PhraseFinder, is proposed to construct collection-dependent association thesauri automatically using large full-text document collections. The association thesaurus can be accessed through natural language queries in INQUERY, an information retrieval system based on the probabilistic inference network. Experiments are conducted in INQUERY to evaluate di erent types of association thesauri, and thesauri constructed for a variety of collections.

• Type: Conference
• Year: 1994
• Where: RIAO
• Authors: Bruce Croft, Jing Yufeng
• Keywords: Association Thesauri, Consistent Bene Ts, Information Retrieval System, RIAO 1994, RIAO 2007