Ontologies for Information Retrieval.

Published in Volume IX, 2000, p. 45-64

Authors: Amalia TODIRASCU

Abstract

The paper presents a system for querying in natural language (French) a set of texts from a limited domain. The domain knowledge, represented in description logics (DL), is used for filtering the results of the search and it is extended dynamically (when new concepts are identified in the texts), as result of DL inference mechanisms. The conceptual hierarchy is built semi-automatically from the texts. Candidate concepts are identified in the texts and checked by the DL tests (satisfiability, consistency etc.). Different small French corpora (heart surgery, newspaper articles, papers on natural language processing) have been used for experimenting the prototype. The system uses shallow natural language parsing techniques and DL reasoning mechanisms are used to handle incomplete or incorrect user queries. A query expansion based on reduced document representation is proposed as an alternative for document retrieval.

BibTeX