Document Retrieval and Routing Using the INQUERY System.

Abstract

eval. We therefore used only two-word phrases. Sequences of more than two words were broken down into two-word subsequences. For example: (crude oil price trend) ) (crude oil), (oil price), (price trend). Using this modification of the original phrase extraction procedures, we were able to eliminate a number of special processing steps, and apply nounphrase -extraction processing to all sections. We weighted the sections of the query as follows; title = 3n; description = 1n, narrative = 1; where n is the number of terms in the section. This was accomplished simply by duplicating the title three times, the description once, and by using the #SUM of the narrative terms. 1.2 PhraseFinder PhraseFinder is a technique for corpus-based query expansion [8]. Phrasal concepts are identified using a part-of-speech tagger, and the text contexts of these concepts (words in close proximity) are used