Systems and Methods Using Query Patterns to Disambiguate Query Intent

• US 20080120276A1
• Filed: 11/16/2006
• Published: 05/22/2008
• Est. Priority Date: 11/16/2006
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Chennavasin, Don
• Granted Patent: US 8,635,203 B2
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/3338 Query expansion
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

Apparatuses, methods, and systems directed to disambiguating queries. Particular embodiments can be used in connection with query analysis and query rewrite processes to determine the intent of one or more keywords contained in a query.

Claims

1. A method, comprising:

accessing a query log to identify one or more patterns that include a selected term, wherein the query log comprises one or more log entries each including a pattern comprising one or more keywords;

determining the frequency of the one or more identified patterns in the query log;

mapping the one or more identified patterns to respective intent categories;

selecting an intent category for the selected term based on the frequencies of the one or more identified patterns.

2. The method of claim 1, further comprising computing a confidence value for the selected intent category.

3. The method of claim 2, wherein the computing step comprises dividing the frequency of the one or more identified patterns that correspond to the selected intent category by an aggregate frequency of the one or more identified patterns.

4. The method of claim 1, wherein the determining step comprises comparing the one or more identified patterns with the one or more log entries in the query log and counting the number of occurrences of the one or more identified patterns.

5. The method of claim 1, wherein the mapping step comprises selecting an intent category for each of the one or more identified patterns.

6. The method of claim 1, wherein the one or more log entries each comprise a time value;

and wherein the identifying step is limited to the log entries having a time value within a threshold period of time.

7. The method of claim 6, wherein the selected intent category for the selected term is associated with the one or more threshold period of time.

8. The method of claim 1 wherein the selected term comprises one or more keywords.

9. A method, for disambiguating the intent of a query, comprising:

receiving a query comprising one or more search terms;

sending the query to a disambiguating engine;

receiving a response from the disambiguating engine, wherein the response comprises one or more intent categories and corresponding confidence values from the disambiguating engine;

selecting a disambiguating experience based on at least one of the confidence values;

presenting search results responsive to the query, wherein one or more aspects of the search results depend on the selected disambiguating experience.

10. The method of claim 9, wherein the response further comprises one or more implicit terms.

11. The method of claim 9 wherein the response comprises one or more suggested search terms.

12. The method of claim 9, wherein the selecting step comprises:

identifying an intent category having the highest confidence value from the one or more results;

selecting a rich experience or a suggestions experience based on the identified highest confidence value and a threshold value.

13. The method of claim 12, wherein the rich experience comprises displaying one or more search results from the identified intent category.

14. The method of claim 13, wherein the displaying step comprises:

selecting the one or more implicit terms corresponding to the identified intent category;

sending the one or more implicit terms with the search query to one or more search engines;

receiving one or more search results from the one or more search engines;

presenting one or more of the received search results.

15. The method of claim 14, wherein the receiving step comprises receiving one or more search results from the one or more search engines and merging the search results.

16. The method of claim 15, wherein the merging step comprises excluding the one or more search results based on one or more threshold values.

17. The method of claim 12, wherein the suggestions experience comprises displaying one or more suggested search terms corresponding to the one or more intent categories.

18. The method of claim 17, wherein the displaying step comprises:

selecting one or more intent categories based on the confidence values of the one or more results;

presenting the one or more suggested search terms from the selected intent categories.

19. A method, comprising:

accessing a query log to identify one or more patterns that include a selected term, wherein the query log comprises one or more log entries each including a pattern comprising one or more keywords;

selecting one or more keywords other than the selected term in the one or more identified patterns;

processing the query log to identify one or more additional patterns that include the selected one or more keywords;

determining the frequency of the one or more identified additional patterns in the query log;

mapping the one or more identified additional patterns to respective intent categories;

selecting an intent category for the selected term based on the frequencies of the one or more additional identified patterns.

20. The method of claim 19, further comprising computing a confidence value for each of the identified additional patterns.

21. The method of claim 20, wherein the computing step comprises dividing the frequency of the one or more identified additional patterns by an aggregate frequency of the one or more identified additional patterns.

22. The method of claim 19, wherein the processing step comprises comparing the one or more identified additional patterns with the one or more log entries in the query log.

23. The method of claim 19, wherein the determining step comprises counting the number of occurrences of the one or more identified additional patterns in the query log.

24. An apparatus, comprising:

a memory;

one or more processors;

logic encoded in one or more computer readable medium, wherein the logic when executed is operable to use the one or more processors to;

access a query log to identify one or more patterns that include a selected term, wherein the query log comprises one or more log entries each including a pattern comprising one or more keywords;

determine the frequency of the one or more identified patterns in the query log;

map the one or more identified patterns to respective intent categories;

select an intent category for the selected term based on the frequencies of the one or more identified patterns.

25. An apparatus, comprising:

a memory;

one or more processors;

logic encoded in one or more computer readable medium, wherein the logic when executed is operable to use the one or more processors to;

receive a query comprising one or more search terms;

send the query to a disambiguating engine;

receive a response from the disambiguating engine, wherein the response comprises one or more intent categories and corresponding confidence values from the disambiguating engine;

select a disambiguating experience based on at least one of the confidence values;

present search results responsive to the query, wherein one or more aspects of the search results depend on the selected disambiguating experience.