Related sentence retrieval system having a plurality of cross-lingual retrieving units that pairs similar sentences based on extracted independent words

A related sentence retrieval system having a first cross-lingual retrieval system for bidirectionally retrieving related sentences between first language sentences and second language sentences, and a second cross-lingual retrieval system for bidirectionally retrieving related sentences between the second language sentences and third language sentences, the system bidirectionally retrieving the related sentences between the first language sentences and the third language sentences by inputting the second language sentences retrieved by one of the two cross-lingual retrieval system to the other system, wherein

• the first cross-lingual retrieval system comprises:
• a first pair data storing unit that stores a plurality of pair data pieces each having a first language sentence and a second language sentence having a same meaning;
• a first language sentence input unit that receives a first language sentence;
• a second language sentence input unit that receives second language sentences from the second cross-lingual retrieval system;
• a first primary retrieving unit that performs retrieval, on the basis of the first language sentence received from the first language sentence input unit or the second language sentences received from the second language sentence input unit, on a set of sentences of the received language stored in the first pair data storing unit;
• a first secondary retrieving unit that performs, based on an independent word extracted from the sentences of the other language paired with the respective sentences retrieved by the first primary retrieving unit, similar sentence retrieval on a set of the sentences of the other language stored in the first pair data storing unit;
• a first language sentence output unit that outputs the first language sentences retrieved by the first secondary retrieving unit;
• and
• a second language sentence output unit that outputs the second language sentences retrieved by the first secondary retrieving unit to the second cross-lingual retrieval system;
• and
• the second cross-lingual retrieval system comprises:
• a second pair data storing unit that stores a plurality of pair data pieces each having a second language sentence and a third language sentence having a same meaning;
• a second language sentence input unit that receives the second language sentences from the first cross-lingual retrieval system;
• a third language sentence input unit that receives a third language sentence;
• a second primary retrieving unit that performs retrieval, on the basis of the second language sentences received from the second language sentence input unit or the third language sentence received from the third language sentence input unit, on a set of sentences of the received language stored in the second pair data storing unit;
• a second secondary retrieving unit that performs, based on an independent word extracted from the sentences of the other language paired with the respective sentences retrieved by the second primary retrieving unit, similar sentence retrieval on a set of the sentences of the other language stored in the second pair data storing unit;
• a third language sentence output unit that outputs the third language sentences retrieved by the second secondary retrieving unit;
• and
• a second language sentence output unit that outputs the second language sentences retrieved by the second secondary retrieving unit to the first cross-lingual retrieval system.

A related sentence retrieval system, having a former stage cross-lingual retrieval system for retrieving, on the basis of a query written in a first language, related sentences written in a second language, and a latter stage cross-lingual retrieval system for retrieving, on the basis of the related sentences in the second language outputted from the former stage cross-lingual retrieval system, sentences written in a third language related to the query written in the first language using the retrieved second language sentences, wherein

• the former stage cross-lingual retrieval system comprises:
• a former stage pair data storing unit that stores a plurality of pair data pieces each having a first language sentence and a second language sentence having a same meaning;
• a first language sentence input unit that receives a query written in the first language;
• a former stage primary retrieving unit that retrieves first language sentences similar to the query from among a set of the first language sentences stored in the former stage pair data storing unit;
• a former stage secondary retrieving unit that performs, based on an independent word extracted from second language sentences paired with the respective first language sentences retrieved by the former primary retrieving unit, similar sentence retrieval from among a set of the second language sentences stored in the former pair data storing unit;
• and
• a second language sentence output unit that outputs the second language sentences retrieved by the former stage secondary retrieving unit;
• and
• the latter stage cross-lingual retrieval system comprises:
• a latter stage pair data storing unit that stores a plurality of pair data pieces each having a second language sentence and a third language sentence having a same meaning;
• a second language sentence input unit that receives sentences in the second language outputted from the second language sentence output unit of the former stage cross-lingual retrieval system;
• a latter stage primary retrieving unit that performs retrieval on the set of the second language sentences stored in the latter stage pair data storing unit on the basis of the received second language sentences;
• a latter stage secondary retrieving unit that performs, based on an independent word extracted from third language sentences paired with the respective second language sentences retrieved by the latter stage primary retrieving unit, similar sentence retrieval from among a set of the third language sentences stored in the latter stage pair data storing unit;
• and
• a third language sentence output unit that outputs the third language sentences retrieved by the latter stage secondary retrieving unit.