Acquiring ontological knowledge from query logs

• US 20080301121A1
• Filed: 05/29/2007
• Published: 12/04/2008
• Est. Priority Date: 05/29/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Suzuki, Hisami, Sekine, Satoshi
• Granted Patent: US 8,051,056 B2
• Time in Patent Office: Days
• US Class Current: 707/5
• CPC Class Codes: G06F 16/367 Ontology

Abstract

Methods are disclosed for acquiring ontological knowledge using query logs. In one embodiment, query logs are first utilized as a basis for identifying important contexts associated with terms belonging to a semantic category. Then, those contexts are as a basis for identifying new terms belonging to the same category or, in another embodiment, as a basis for removing extraneous or obsolete terms identified as being in the same category.

Claims

1. A computer-implemented method for acquiring ontological knowledge, the method comprising:

obtaining a seed term that is classified within a semantic category; and

identifying, from a query log, a context associated with the seed term.

2. The method of claim 1, wherein identifying comprises identifying a word that appears within a query within the query log in a position that is proximate to the seed term.

3. The method of claim 1, further comprising identifying, from a query within the query log, a new term for classification within the semantic category.

4. The method of claim 1, further comprising identifying, based at least in part on the context, a term that should be removed from the semantic category.

5. The method of claim 1, further comprising determining whether an appearance of the context with a different term is likely to be indicative of whether the different term belongs to the semantic category.

6. The method of claim 5, wherein determining comprises evaluating the frequency with which the context occurs within the query log.

7. The method of claim 6, further comprising biasing identification of the context away from a context that occurs relatively frequently within the query log.

8. A computer-implemented method for acquiring ontological knowledge, the method comprising identifying, from a query within a query log, a new term for classification within a semantic category.

9. The method of claim 8, wherein identifying a new term for classification within the semantic category comprises identifying based on words that appear with the new term within the query.

10. The method of claim 8, wherein identifying a new term for classification within the semantic category comprises identifying based on words that appear within the query log with a term that is known to fit within the semantic category.

11. The method of claim 8, wherein identifying a new term for classification within the semantic category comprises identifying based on analysis of information contained in the query log.

12. The method of claim 8, wherein identifying a new term for classification within a semantic category comprises identifying a new term for classification within a user-selected semantic category.

13. The method of claim 8, wherein identifying a new term for classification within a semantic category comprises identifying a new term for classification based at least in part on a seed term received from a user.

14. The method of claim 8, wherein the semantic category is related to a person.

15. The method of claim 8, wherein the semantic category is related to a location.

16. The method of claim 8, wherein the semantic category is related to a physical object.

17. The method of claim 8, wherein the semantic category is customizable.

18. A computer-implemented method for acquiring ontological knowledge, the method comprising:

receiving an indication of a category and a seed term;

identifying at least one instance of the seed word within a query log; and

utilizing said at least one instance as a basis for identifying a term that occurs within the query log and is consistent with the category.

19. The method of claim 18, wherein receiving an indication of a category comprises receiving a user-initiated input.

20. The method of claim 18, wherein receiving a user-initiated input comprises receiving an indication of a custom category.