Cross-lingual retrieval system and method that utilizes stored pair data in a vector space model to process queries

Abstract

The cross-lingual retrieval system of the present invention retrieves second language sentences which are appropriate translations of a query written in a first language using pair data without being greatly influenced by the difference in expression between the query and sentences to be the object of retrieval, length of the query or sentences to be the object of retrieval or the like. In the system, a pair data storing unit stores multiple pairs each having a first language sentence and a second language sentence having the same meaning. As a query receiving unit receives a query written in the first language, a first retrieving unit retrieves first language sentences similar to the query from a set of first language sentences stored in the pair data storing unit. A second retrieving unit then retrieves second language sentences similar to second language sentences paired with the respective first language sentences retrieved by the first retrieving unit from a set of second language sentences stored in the pair data storing unit. That is, a first retrieval of first language sentences from the pair data is performed based on a query, and then a second retrieval of second language sentences from the pair data is performed based on the result of the first retrieval.