An Approach to Conceptual Text Retrieval Using the EuroWordNet Multilingual Semantic Database

This paper explores the application of a multilingual, WordNet-like lexical knowledge base, EuroWordNet, to Cross-Language Text Retrieval. EuroWordNet (EWN) is a multilingual database with basic semantic relations between words for some European languages (Dutch, Italian, Spanish and English). In addition to the relations in WordNet 1.5, EWN includes domain labels, cross-language and cross-category relations, which are directly useful for Multilingual Information Retrieval. We propose conceptual, language-independent indexing on the basis of EWN Interlingual Index (ILI) as a promising approach to cross-language text retrieval. Introduction There is a general consensus in the domain of multilingual text retrieval (MTR) about the necessity to develop large-scale resources in order to improve the capacities and performance of current MTR systems (Fluhr 1996; Oard & Dorr 1996). In particular, largescale databases storing basic semantic relations between words can be used for reformulatio..