Systems and methods for using anchor text as parallel corpora for cross-language information retrieval

• US 8,631,010 B1
• Filed: 05/18/2012
• Issued: 01/14/2014
• Est. Priority Date: 08/28/2001
• Status: Expired due to Fees

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Gravano, Luis, Henzinger, Monika H.
• Primary Examiner(s): Jalil, Neveen Abel
• Assistant Examiner(s): Veillard, Jacques
• Application Number:
  US13/474,957
• Time in Patent Office: 606 Days
• Field of Search: None
• US Class Current: 707/736
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9532 Query formulation
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access

Abstract

A method may include obtaining, based on a content of a search query, one or more documents in a first language; identifying one or more documents in a second language that contain an anchor that links to the one or more documents in the first language, the second language being different than the first language; and translating one or more terms of the search query into the second language using content included in the one or more documents in the second language.

Claims

1. A method, performed by one or more processors, the method comprising:

receiving, by the one or more processors, a search query in a first natural language;

identifying, by the one or more processors and based on a content of the search query, one or more documents in the first natural language;

identifying, by the one or more processors, one or more documents in a second natural language that contain an anchor link that links to the one or more documents in the first natural language,

the second natural language being different than the first natural language;

analyzing, by the one or more processors, content included in the one or more documents in the second natural language; and

translating, by the one or more processors and based on analyzing the content included in the one or more documents in the second natural language, one or more terms of the search query into the second natural language.

2. The method of claim 1, where translating the one or more terms of the search query comprises:

performing an initial translation of the one or more terms of the search query, into the second natural language, to generate possible translations of the one or more terms of the search query, and

disambiguating among the possible translations of the one or more terms of the search query using the content, included in the one or more documents in the second natural language, to identify one of the possible translations as a likely translation of the one or more terms of the search query.

3. The method of claim 2, where performing the initial translation of the one or more terms of the search query comprises:

using a dictionary to identify the possible translations of the one or more terms in the second natural language.

4. The method of claim 2, where disambiguating among the possible translations of the one or more terms of the search query comprises:

using text from the one or more documents in the second natural language as parallel corpora to disambiguate among the possible translations of the one or more terms of the search query.

5. The method of claim 1, where the content included in the one or more documents in the second natural language includes text associated with the anchor link.

6. The method of claim 1, where the content included in the one or more documents in the second natural language includes an entirety of text of the one or more documents in the second natural language.

7. The method of claim 1, where identifying the one or more documents in the first natural language comprises:

performing a search for documents in the first natural language using the search query in the first natural language, and

identifying, based on performing the search, the one or more documents in the first natural language.

8. A system comprising:

one or more processors to;

receive a search query in a first natural language;

identify, based on a content of the search query, one or more documents in the first natural language;

identify one or more documents in a second natural language that contain an anchor link that links to the one or more documents in the first natural language,

the second natural language being different than the first natural language;

analyze content included in the one or more documents in the second natural language; and

translate, based on analyzing the content included in the one or more documents in the second natural language, one or more terms of the search query into the second natural language.

9. The system of claim 8, where, when translating the one or more terms of the search query, the one or more processors are further to:

perform an initial translation of the one or more terms of the search query, into the second natural language, to generate possible translations of the one or more terms of the search query, and

disambiguate among the possible translations of the one or more terms of the search query using the content, included in the one or more documents in the second natural language, to identify one of the possible translations as a likely translation of the one or more terms of the search query.

10. The system of claim 9, where, when performing the initial translation of the one or more terms of the search query, the one or more processors are further to:

use a dictionary to identify the possible translations of the one or more terms in the second natural language.

11. The system of claim 9, where, when disambiguating among the possible translations of the one or more terms of the search query, the one or more processors are further to:

use text from the one or more documents in the second natural language as parallel corpora to disambiguate among the possible translations of the one or more terms of the search query.

12. The system of claim 8, where the content included in the one or more documents in the second natural language includes text associated with the anchor link.

13. The system of claim 8, where the content included in the one or more documents in the second natural language includes an entirety of text of the one or more documents in the second natural language.

14. The system of claim 8, where, when identifying the one or more documents in the first natural language, the one or more processors are further to:

perform a search for documents in the first natural language using the search query in the first natural language, and

identify, based on performing the search, the one or more documents in the first natural language.

15. A non-transitory computer-readable medium containing instructions for execution by one or more processors, the instructions comprising:

one or more instructions that, when executed by the one or more processors, cause the one or more processors to;

receive a search query in a first natural language;

identify, based on a content of the search query, one or more documents in the first natural language;

identify one or more documents in a second natural language based on an anchor link, contained in the one or more documents in the first natural language, that links the one or more documents in the first natural language to the one or more documents in the second natural language,

the second natural language being different than the first natural language;

analyze content included in the one or more documents in the second natural language; and

translate, based on analyzing the content included in the one or more documents in the second natural language, one or more terms of the search query into the second natural language.

16. The computer-readable medium of claim 15, where the one more instructions that cause the one or more processors to translate the one or more terms of the search query include:

one or more instructions to cause the one or more processors to;

perform an initial translation of the one or more terms of the search query, into the second natural language, to generate possible translations of the one or more terms of the search query, and

disambiguate among the possible translations of the one or more terms of the search query using the content, included in the one or more documents in the second natural language, to identify one of the possible translations as a likely translation of the one or more terms of the search query.

17. The computer-readable medium of claim 16, where the one more instructions that cause the one or more processors to disambiguate among the possible translations of the one or more terms of the search query include:

one or more instructions to cause the one or more processors to use text from the one or more documents in the second natural language as parallel corpora to disambiguate among the possible translations of the one or more terms of the search query.

18. The computer-readable medium of claim 15, where the content included in the one or more documents in the second natural language includes text associated with the anchor link.

19. The computer-readable medium of claim 15, where the content included in the one or more documents in the second natural language includes an entirety of text of the one or more documents in the second natural language.

20. The computer-readable medium of claim 15, where the one more instructions that cause the one or more processors to identify the one or more documents in the first natural language include:

one or more instructions to cause the one or more processors to;

perform a search for documents in the first natural language using the search query in the first natural language, and

identify, based on performing the search, the one or more documents in the first natural language.