Using Information Extraction to Improve Document Retrieval

An information extraction system was adapted to act as a post-filter on the output of an IR system to improve precision on routing tasks and make it easier to write IE grammars for multiple topics.

Abstract

We describe an approach to applying a particular kind of Natural Language Processing NLP system to the TREC routing task in Information Retrieval IR Rather than attempting to use NLP techniques in indexing documents in a corpus we adapted an information extraction IE system to act as a postlter on the output of an IR system The IE system was congured to score each of the top \t\t\t documents as determined by an IR system and on the basis of that score to rerank those \t\t\t documents One aim was to improve precision on routing tasks Another was to make it easier to write IE grammars for multiple topics Researchers have pursued a variety of approaches to integrating natural lan guage processing with document retrieval systems The central idea in the liter ature is that some perhaps shallow variant of the kind of syntactic and semantic analysis performed by generalpurpose natural language processing systems can