Exploiting lexical resources and linguistic tools in cross-language information retrieval: the euro-search approach

Cross-language Information Retrieval (CLIR) is a very new research area in which methodologies and tools developed for Natural Language Processing (NLP) are being integrated with techniques and results coming from the Information Retrieval (IR) field. The EuroSearch project provides an excellent testbed for the application and testing of different kinds of cross-language retrieval methodologies. In EuroSearch, a federation of industrial search engines (Arianna for Italy, Ole for Spain and EuroSpider for Switzerland), using different kinds of search mechanisms, will provide their users with functionalities for cross-language querying. The paper describes and motivates the technology that will be adopted in the implementation of the multilingual interface of the federation, focusing on an approach that enhances the potential of a lexicon-based search through the integration of a corpus-based methodology.

Sommario non disponibile.