Abstract.
The paper presents an Information Retrieval (IR) system for a limited domain providing natural language interfaces. 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. Concept instances are identified using shallow natural language parsing techniques. The paper investigates also various possibilities of integrating semantic knowledge into an IR system.