Cross-lingual indexing and information retrieval

• US 8,135,575 B1
• Filed: 08/21/2003
• Issued: 03/13/2012
• Est. Priority Date: 08/21/2003
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Dean, Jeffrey A.
• Primary Examiner(s): Shah, Paras
• Application Number:
  US10/647,053
• Time in Patent Office: 3,127 Days
• Field of Search: 704 1- 10, 707/3, 707/4, 707/705, 707/706, 707/708, 715/256, 715/264
• US Class Current: 704/7
• CPC Class Codes: G06F 16/243 Natural language query formulation
• CPC Class Codes: G06F 40/49 using very large corpora, e.g. the web
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation

Abstract

Systems and methods are disclosed for searching across multi-lingual information. A user makes a query in a first language, and a group of documents that were previously machine-translated into the first language are searched for information responsive to the query. Contextual information derived can be used to improve the accuracy of the machine translation. Responsive documents are returned to the user. Alternatively, a query provided in a user'"'"'s language may be translated into one or more other languages. Documents written in these languages can then be searched for information responsive to the appropriate translated query. Responsive documents can be translated into the user'"'"'s language prior to providing them to the user.

Claims

1. A system comprising:

a document database, the document database containing a group of documents, each document having been translated into multiple languages using a context-specific translation model where the context-specific translation model uses, at least in part, documents each having a link that points to a document to be translated;

a database of anchor text for links to a plurality of the documents in the document database, the database of anchor text having been translated into multiple languages using the context-specific translation model; and

one or more computers capable of executing a program which, when executed by the computers, performs operations comprising receiving a query in a given language and searching documents in the document database and the anchor text that are written in the given language, or that have been translated into the given language, for information responsive to the query, the given language being one of the multiple languages.

2. The system of claim 1, further comprising:

a machine translation engine for automatically translating the document into multiple languages.

3. The system of claim 2, where the machine translation engine is located remotely from the document database.

4. The system of claim 1, where the document database further comprises anchor text associated with at least some of said documents, and where the operations further comprise searching the document database anchor text for information responsive to the query.

5. The system of claim 1, where the plurality of documents are obtained from web pages and offline documents.

6. A computer-implemented method comprising:

creating a document database, the document database containing a group of documents, each document having been translated into multiple languages using a context-specific translation model where the context-specific translation model uses, at least in part, documents each having a link that points to a document to be translated;

creating a database of anchor text for links to a plurality of the documents in the document database, the database of anchor text having been translated into multiple languages using the context-specific translation model; and

receiving a query in a given language and searching documents in the document database and the anchor text that are written in the given language, or that have been translated into the given language, for information responsive to the query, the given language being one of the multiple languages;

where the creating and receiving are performed by one or more processors.

7. The method of claim 6, further comprising translating the document into multiple languages.

8. The method of claim 6, where the document database further comprises anchor text associated with at least some of said documents, the method further comprising searching the document database anchor text for information responsive to the query.

9. The method of claim 6 where the plurality of documents are obtained from web pages and offline documents.

10. A non-transitory computer readable media encoded with instructions that are operable, when executed by a computer, to cause the computer to perform operations comprising:

creating a document database, the document database containing a group of documents, each document having been translated into multiple languages using a context-specific translation model where the context-specific translation model uses, at least in part, documents each having a link that points to a document to be translated;

creating a database of anchor text for links to a plurality of the documents in the document database, the database of anchor text having been translated into multiple languages using the context-specific translation model; and

receiving a query in a given language and searching documents in the document database and the anchor text that are written in the given language, or that have been translated into the given language, for information responsive to the query, the given language being one of the multiple languages.

11. The non-transitory computer readable media of claim 10, where the operations further comprise translating the document into multiple languages.

12. The non-transitory computer readable media of claim 10, where the document database further comprises anchor text associated with at least some of said documents, where the operations further comprise searching the document database anchor text for information responsive to the query.

13. The non-transitory computer readable media of claim 10 where the plurality of documents are obtained from web pages and offline documents.