LLaJú: Multilingual Information Retrieval

Description

LlaJú is a Multilingual Information Retrieval system based on statistical techniques. The demonstration implemented here is designed to operate on five languages belonging to the European Community, from a query necessarily expressed in the English language. The search will be carried out on documents written in German, Spanish, French, English and Italian. The indexed corpus is composed of more than one million news items published in 1994 in various newspapers and newswire agencies (Der Spiegel, EFE, Le Mode, Los Angeles Times, La Stampa, etc). The queries should be in English and will be translated into different languages.