DETERMINING CORRESPONDING TERMS WRITTEN IN DIFFERENT FORMATS

• US 20140188454A1
• Filed: 03/06/2014
• Published: 07/03/2014
• Est. Priority Date: 07/06/2000
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Mittal, Vibhu, Ponte, Jay M., Sahami, Mehran, Ghemawat, Sanjay, Bauer, John A.
• Granted Patent: US 9,734,197 B2
• Time in Patent Office: Days
• US Class Current: 704/2
• CPC Class Codes: G06F 16/2452 Query translation
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries
• CPC Class Codes: G06F 16/3332 Query translation
• CPC Class Codes: G06F 16/972 Access to data in other repository systems, e.g. legacy data or dynamic Web page generation
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 40/12 Use of codes for handling textual entities
• CPC Class Codes: G06F 40/237 Lexical tools

Abstract

Methods and apparatus consistent with the invention allow a user to submit an ambiguous search query and to receive relevant search results. Queries can be expressed using character sets and/or languages that are different from the character set and/or language of at least some of the data that is to be searched. A translation between these character sets and/or languages can be performed by examining the use of terms in aligned text. Probabilities can be associated with each possible translation. Refinements can be made to these probabilities by examining user interactions with the search results.

Claims

1. A method comprising:

identifying a first set of anchor text written in a first format and containing a given term;

identifying a set of documents to which the first set of anchor text points;

identifying a second set of anchor text written in a second format and pointing to the identified set of documents;

analyzing the second set of anchor text to determine that a representation of the given term in the first format corresponds to a representation of the given term in the second format.

2. The method of claim 1, wherein the first format comprises a first character set, and the second format comprises a second character set.

3. The method of claim 2, wherein the first format is selected from the group consisting of:

romaji, romaja, and pinyin; and

in which the second character set is selected from the group consisting of;

katakana, hiragana, kanji, hangul, hanja, and traditional Chinese characters.

4. The method of claim 1, wherein the first format comprises a first language and the second format comprises a second language.

5. The method of claim 1, wherein analyzing the second set of anchor text includes identifying a term that appears most frequently in the second set of anchor text and designating the most frequently appearing term as the representation of the given term in the second format.

6. The method of claim 1, wherein analyzing the second set of anchor text comprises:

calculating a probability that the given term corresponds to a term in the second set of anchor text.

7. The method of claim 6, wherein the probability is obtained using at least one of Bayesian methods, histogram smoothing, kernel smoothing, and shrinkage estimators.

8. The method of claim 6, wherein the probability that the given term corresponds to a term in the second set of anchor text is obtained by dividing the number of occurrences of the term in the second set of anchor text by the total number of occurrences of all terms in the second set of anchor text.

9. The method of claim 1, wherein analyzing the second set of anchor text comprises:

calculating a probability that the given term corresponds to each term in the second set of anchor text.

10. The method of claim 1, wherein analyzing the second set of anchor text comprises:

identifying a term that appears most frequently in the second set of anchor text.

11. The method of claim 1, wherein the documents comprise web pages.

12. The method of claim 1, further comprising:

obtaining a query written in the first format and containing the given term;

translating the query into the second format based at least in part on said analyzing step;

searching a database for information written in the second format that is responsive to the translated query.

13. The method of claim 12, wherein the steps are performed in the order recited.

14. A computer program product embodied on a non-transitory computer-readable medium, the computer program product including instructions, which when executed by a computer system, are operable to cause the computer system to perform operations comprising:

identifying a first set of anchor text written in a first format and containing a given term;

identifying a set of web pages to which the first set of anchor text points;

identifying a second set of anchor text written in a second format and pointing to the identified set of web pages;

determining a probability that a representation of the given term in the first format corresponds to a representation of the given term in the second format.

15. The computer program product of claim 14, wherein the operations further comprise:

modifying the probability that a representation of the given term in the first format corresponds to a representation of the given term in the second format based, at least in part, on an analysis of a user'"'"'s selection of search results.

16. The computer program product of claim 14, wherein the operations further comprise:

modifying the probability that a representation of the given term in the first format corresponds to a representation of the given term in the second format based, at least in part, on an analysis of a user'"'"'s previous queries.

17. The computer program product of claim 14, wherein the probability is determined, at least in part, using at least one of Bayesian methods, histogram smoothing, kernel smoothing, and shrinkage estimators.