Multilingual document retrieval system

• US 7,047,182 B2
• Filed: 12/13/2001
• Issued: 05/16/2006
• Est. Priority Date: 12/20/2000
• Status: Expired due to Fees

• Current Assignee: Fuji Xerox Company Limited (Fujifilm Holdings Corporation)
• Original Assignee: Fuji Xerox Company Limited (Fujifilm Holdings Corporation)
• Inventors: Masuichi, Hiroshi
• Primary Examiner(s): Young, W. R.
• Assistant Examiner(s): Sked, Matthew J.
• Application Number:
  US10/013,372
• Publication Number:
  US 20020123982A1
• Time in Patent Office: 1,615 Days
• Field of Search: None
• US Class Current: 704/7
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/55 Rule-based translation
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

As a retrieval result, appropriate text of a second language is provided in response to a retrieval request by text of a first language. A first directory storing part stores a first directory structure created for a first language. A second directory storing part stores a second directory structure created for a second language. A directory relation storing part stores correspondences between directories in the first directory structure and directories in the second directory structure. A directory retrieval part receives a retrieval request by the first language from a user and decides which directory in the first directory structure the request has a high degree of relation with. A multilingual retrieval part decides documents having a high degree of relation with the retrieval request, of documents belonging to a directory in the second directory structure that corresponds to the decided directory.

Claims

1. A multilingual document retrieval system, comprising:

a first directory storing unit that stores a first directory structure created for a first language;

a second directory storing unit that stores a second directory structure created for a second language;

a directory relation storing unit that stores a correspondence between a directory in the first directory structure and a directory in the second directory structure;

a directory retrieval unit that decides which directory in the first directory structure has a high degree of relation with a retrieval request by the first language from a user, and the decision includes information related to levels within the first directory structure; and

a multilingual retrieval unit that decides a document having a high degree of relation with the retrieval request by the first language from the user among documents belonging to a directory in the second directory structure that corresponds to the directory decided by the directory retrieval unit.

2. The multilingual document retrieval system according to claim 1, comprising:

a learning data storing unit that stores, for each of pairs of corresponding directories stored in the directory relation storing unit, multilingual retrieval learning data that includes dictionary data of a field of the directory pair and a translation pair,

wherein the multilingual retrieval unit decides a document having a high degree of relation with the retrieval request by the first language from the user among documents belonging to a directory in the second directory structure that corresponds to the directory decided by the directory retrieval unit, using corresponding learning data stored in the learning data storing unit.

3. The multilingual document retrieval system according to claim 1, comprising:

a learning data storing unit that extracts a pair of a text by the first language and a text by the second language from a document having the text pair among documents belonging to the pair of corresponding directories stored in the directory relation storing unit, and stores the text pair as multilingual retrieval learning data,

wherein the multilingual retrieval unit decides a document having a high degree of relation with the retrieval request by the first language from the user among documents belonging to a directory in the second directory structure that corresponds to the directory decided by the directory retrieval unit, using corresponding learning data stored in the learning data storing unit.

4. A multilingual document retrieval system, comprising:

a first directory storing unit that stores a first directory structure created for a first language;

a second directory storing unit that stores a second directory structure created for a second language;

a directory relation storing unit that stores a correspondence between a directory in the first directory structure and a directory in the second directory structure;

a multilingual retrieval unit that retrieves a pair of a document by the first language and a document by the second language that have a relation with each other from the pair of corresponding directories stored in the directory relation storing unit, and a retrieval decision includes information related to levels within the first directory structure or the second directory structure; and

a retrieval unit that decides which document in the first directory structure has a relation with a retrieval request by the first language from a user.

5. The multilingual document retrieval system according to claim 4, comprising:

a learning data storing unit that stores, for each of pairs of corresponding directories stored in the directory relation storing unit, multilingual retrieval learning data that includes dictionary data of a field of the directory pair and a translation pair,

wherein the multilingual retrieval unit retrieves the pair of document by the first language and document by the second language that have a relation with each other from the pair of corresponding directories stored in the directory relation storing unit, using corresponding learning data stored in the learning data storing unit.

6. The multilingual document retrieval system according to claim 4, comprising:

a learning data storing unit that extracts a pair of a text by the first language and a text by the second language from a document having the text pair among documents belonging to the pair of corresponding directories stored in the directory relation storing unit, and stores the text pair as multilingual retrieval learning data,

wherein the multilingual retrieval unit retrieves the pair of document by the first language and document by the second language that have a relation with each other from the pair of corresponding directories stored in the directory relation storing unit, using corresponding learning data stored in the learning data storing unit.

7. A multilingual document retrieval system, comprising:

a first directory storing unit that stores a first directory structure created for a first language;

a second directory storing unit that stores a second directory structure created for a second language;

a directory relation storing unit that stores a correspondence between a directory in the first directory structure and a directory in the second directory structure;

a directory retrieval unit that decides which directory in the first directory structure has a high degree of relation with a retrieval request by the first language from a user, and the decision includes information related to levels within the first directory structure;

a translation unit that translates the retrieval request by the first language from the user to a retrieval request by the second language; and

a retrieval unit that decides a document having a high degree of relation with the retrieval request by the second language obtained by the translation unit among documents belonging to a directory in the second directory structure that corresponds to the directory decided by the directory retrieval unit.

8. The multilingual document retrieval system according to claim 7, comprising:

a learning data storing unit that stores, for each of pairs of corresponding directories stored in the directory relation storing unit, multilingual retrieval learning data that includes dictionary data of a field of the directory pair and a translation pair,

wherein the translation unit translates the retrieval request by the first language from the user to a retrieval request by the second language, using the learning data stored in the learning data storing unit, the learning data corresponding to the directory obtained by the directory retrieval unit.

9. The multilingual document retrieval system according to claim 7, comprising:

a learning data storing unit that extracts a pair of a text by the first language and a text by the second language from a document having the text pair among documents belonging to the pair of corresponding directories stored in the directory relation storing unit, and stores the text pair as multilingual retrieval learning data,

wherein the translation unit translates the retrieval request by the first language from the user to a retrieval request by the second language, using the learning data stored in the learning data storing unit, the learning data corresponding to the directory obtained by the directory retrieval unit.

10. A multilingual document retrieval system, comprising:

a first directory storing unit that stores a first directory structure created for a first language;

a second directory storing unit that stores a second directory structure created for a second language;

a directory relation storing unit that stores a correspondence between a directory in the first directory structure and a directory in the second directory structure;

a directory retrieval unit that decides which directory in the first directory structure has a high degree of relation with a retrieval request by the first language from a user, and the decision includes information related to levels within the first directory structure;

a translation unit that translates a document by the second language in the second directory structure to a document by the first language; and

a retrieval unit that decides a document having a high degree of relation with the retrieval request by the first language from the user among documents translated to the first language by the translation unit that belong to a directory in the second directory structure corresponding to the directory decided by the directory retrieval unit.

11. The multilingual document retrieval system according to claim 10, comprising:

a learning data storing unit that stores, for each of pairs of corresponding directories stored in the directory relation storing unit, multilingual retrieval learning data that includes dictionary data of a field of the directory pair and a translation pair,

wherein the translation unit translates the document by the second language in the second directory structure to a document by the first language using the learning data stored in the learning data storing unit, the learning data corresponding to the directory.

12. The multilingual document retrieval system according to claim 10, comprising:

a learning data storing unit that extracts a pair of a text by the first language and a text by the second language from a document having the text pair among documents belonging to the pair of corresponding directories stored in the directory relation storing unit, and stores the text pair as multilingual retrieval learning data,

wherein the translation unit translates the document by the second language in the second directory structure to a document by the first language using the learning data stored in the learning data storing unit, the learning data corresponding to the directory.

13. A parallel corpus creating system, comprising:

a first directory storing unit that stores a first directory structure created for a first language;

a second directory storing unit that stores a second directory structure created for a second language;

a directory relation storing unit that stores a correspondence between a directory in the first directory structure and a directory in the second directory structure; and

a multilingual retrieval unit that retrieves a pair of a document by the first language and a document by the second language that have a relation with each other from the pair of corresponding directories stored in the directory relation storing unit, and retrieval decision includes information related to levels within the first or second directory structure.

14. A parallel corpus creating system, comprising:

a first directory storing unit that stores a first directory structure created for a first language;

a second directory storing unit that stores a second directory structure created for a second language;

a directory relation storing unit that stores a correspondence between a directory in the first directory structure and a directory in the second directory structure;

a learning data storing unit that stores, for each of pairs of corresponding directories stored in the directory relation storing unit, multilingual retrieval learning data that includes dictionary data of a field of the directory pair and a translation pair; and

a multilingual retrieval unit that retrieves a pair of a document by the first language and a document by the second language that have a relation with each other from the pair of corresponding directories stored in the directory relation storing unit, using corresponding learning data stored in the learning data storing unit and information related to levels within the first or the second directory structure.

15. A parallel corpus creating system, comprising:

a first directory storing unit that stores a first directory structure created for a first language;

a second directory storing unit that stores a second directory structure created for a second language;

a directory relation storing unit that stores a correspondence between a directory in the first directory structure and a directory in the second directory structure;

a learning data storing unit that extracts a pair of a text by the first language and a text by the second language from a document having the text pair among documents belonging to the pair of corresponding directories stored in the directory relation storing unit, and stores the text pair as multilingual retrieval learning data; and

a multilingual retrieval unit that retrieves the pair of document by the first language and document by the second language that have a relation with each other from the pair of corresponding directories stored in the directory relation storing unit, using corresponding learning data stored in the learning data storing unit and information related to levels within the first or the second directory structure.

16. A multilingual document retrieval system, comprising:

a directory relation storing unit that stores a correspondence between a directory in a first directory structure and a directory in a second directory structure;

a directory deciding unit that decides a directory in the second directory structure corresponding to a directory in the first directory structure that has a high degree of relation with a retrieval request from a user on the directory correspondence stored in the directory relation storing unit; and

a retrieval request issuing unit that issues a retrieval request to retrieve a document having a high degree of relation with the retrieval request from the user among documents belonging to the decided directory in the second directory structure.

17. A multilingual document retrieval system, comprising:

a directory relation storing unit that stores a correspondence between a directory in a first directory structure created for a first language and a directory in a second directory structure created for a second language;

a directory deciding unit that decides a directory in the second directory structure corresponding to a directory in the first directory structure that has a high degree of relation with a retrieval request by the first language from a user based on the directory correspondence stored in the directory relation storing unit, and the decision includes information related to levels within the first directory structure; and

a retrieval request issuing unit that issues a retrieval request to retrieve a document having a high degree of relation with the retrieval request by the first language from the user among documents belonging to the decided directory in the second directory structure.

18. A multilingual document retrieval system, comprising:

a directory relation storing unit that stores a correspondence between a directory in a first directory structure created for a first language and a directory in a second directory structure created for a second language;

a translation unit that translates a retrieval request by the first language from a user to a retrieval request by the second language;

a directory deciding unit that decides a directory in the second directory structure corresponding to a directory in the first directory structure that has a high degree of relation with the retrieval request by the first language from the user based on the directory correspondence stored in the directory relation storing unit; and

a retrieval request issuing unit that issues a retrieval request to retrieve a document having a high degree of relation with the retrieval request from the user translated from the first language to the second language among documents belonging to the decided directory in the second directory structure.

19. A multilingual document retrieval system, comprising:

a first directory storing unit that stores a first directory structure created for a first language;

a second directory storing unit that stores a second directory structure created for a second language;

a communication unit that communicates with the second directory storing unit;

a directory relation storing unit that stores a correspondence between a directory in the first directory structure and a directory in the second directory structure;

a directory retrieval unit that decides a directory in the first directory structure in response to a retrieval request by the first language from a user, and the decision includes information related to levels within the first directory structure; and

a multilingual retrieval unit that decides a document having a high degree of relation with the retrieval request by the first language from the user among documents belonging to a directory in the second directory structure that corresponds to the directory decided by the directory retrieval unit via the communication unit.

20. A multilingual document retrieval method, comprising:

storing a first directory structure created for a first language;

storing a second directory structure created for a second language;

storing a correspondence between a directory in the first directory structure and a directory in the second directory structure;

deciding which directory in the stored first directory structure has a high degree of relation with a retrieval request by the first language from a user, and the decision includes information related to levels within the first directory structure; and

deciding, by using the stored correspondence, a document having a high degree of relation with the retrieval request by the first language from the user among documents belonging to a stored directory in the stored second directory structure that corresponds to the decided directory.

21. A storage medium readable by a computer, the storage medium storing a program of instructions executable by the computer to perform a function for performing multilingual document retrieval, the function comprising the steps of:

storing a first directory structure created for a first language;

storing a second directory structure created for a second language;

storing a correspondence between a directory in the first directory structure and a directory in the second directory structure;

deciding which directory in the stored first directory structure has a high degree of relation with a retrieval request by the first language from a user, and the decision includes information related to levels within the first directory structure; and

deciding, by using the stored correspondence, a document having a high degree of relation with the retrieval request by the first language from the user among documents belonging to a directory in the stored second directory structure that corresponds to the decided directory.