Automated analysis of user search behavior

• US 20080281809A1
• Filed: 05/10/2007
• Published: 11/13/2008
• Est. Priority Date: 05/10/2007
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Wang, Robert Ching-Huei, Anderson, Blake Edward
• Granted Patent: US 8,037,042 B2
• Time in Patent Office: Days
• US Class Current: 707/5
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

Automated analysis of user search behavior is provided. Data on user searches is maintained in a user search database. Relevance factors are determined for each search result included in a given search session where the relevance factors provide an indication of user satisfaction with particular search results included in the session. The relevance factors for each search result are analyzed by a relevance classification module for classifying each search result in terms of its relevance to an associated search query. The result of the relevance classification may assign a relevance classification and associated confidence level to each analyzed search result as to whether the search result is acceptable, unacceptable or partially acceptable relative to the search query that resulted in the search result. Relevance classifications for each analyzed search result may be stored for future use, for example, for diagnostic analysis of the operation of a given search mechanism.

Claims

1. A method of analyzing search behavior, comprising:

storing data associated with a search and use of a plurality of search results;

identifying one or more of the plurality of search results as a session of related search results;

calculating one or more relevance factors for each search result included in the session of search results; and

providing a relevance classification for each search result included in the session of search results based on a relevance of each search result to a search query that was used to locate each search result.

2. The method of claim 1, prior to calculating one or more relevance factors for each search result included in a session of search results, processing the data to determine a search query that was used to locate each search result.

3. The method of claim 2, prior to calculating one or more relevance factors for each search result included in a session of search results, processing the data to determine click-through behavior for each search result.

4. The method of claim 3, prior to calculating one or more relevance factors for each search result included in a session of search results, processing the data to obtain explicit feedback on each search result.

5. The method of claim 1, wherein identifying one or more of the plurality of search results as a session of related search results includes identifying one or more of the plurality of search results obtained in a prescribed time period.

6. The method of claim 1, wherein providing a relevance classification for each search result included in the session of search results based on a relevance of each search result to a search query that was used to locate each search result includes providing the relevance classification for each search result based on the one or more relevance factors calculated for each search result included in the session of search results.

7. The method of claim 6, wherein providing the relevance classification for each search result based on the one or more relevance factors calculated for each search result included in the session of search results, analyzing user behavior associated with a plurality of search results for which explicit feedback is provided.

8. The method of claim 7, wherein analyzing user behavior associated with a plurality of search results for which explicit feedback is provided includes:

determining a first user behavior profile based on one or more relevance factors associated with a search result for which explicit feedback is provided and for which the search result is acceptable to a searching user;

determining a second user behavior profile based on one or more relevance factors associated with a search result for which explicit feedback is provided and for which the search result is not acceptable to a searching user; and

determining a third user behavior profile based on one or more relevance factors associated with a search result for which explicit feedback is provided and for which the search result is partially acceptable to a searching user.

9. The method of claim 8, wherein providing a relevance classification for each search result included in the session of search results based on a relevance of each search result to a search query that was used to locate each search result includes applying the first, second and third user behavior profiles to one or more relevance factors calculated for each search result included in the session of search results for classifying each search result included in the session of search results for which no explicit feedback is provided as acceptable, unacceptable or partially acceptable.

10. The method of claim 9, wherein if a user behavior associated with a given search result for which no explicit feedback is provided matches a user behavior associated with the first use behavior profile, classifying the given search result as acceptable.

11. The method of claim 10, wherein if a user behavior associated with the given search result for which no explicit feedback is provided matches a user behavior associated with the second user behavior profile, classifying the given search result as unacceptable.

12. The method of claim 11, wherein if a user behavior associated with the given search result for which no explicit feedback is provided matches a user behavior associated with the third user behavior profile, classifying the given search result as partially acceptable.

13. The method of claim 12, further comprising storing the relevance classification for each search result.

14. A system for analyzing search behavior, comprising:

a user search database operative to store data associated with a search and use of a plurality of search results;

a session identifier module operative to identify one or more of the plurality of search results as a session of related search results;

a relevance calculation module operative to calculate one or more relevance factors for each search result included in the session of search results; and

a relevance classifier module operative to provide a relevance classification for each search result included in the session of search results based on a relevance of each search result to a search query that was used to locate each search result.

15. The system of claim 14, further comprising:

a query processor module operative to process the data to determine a search query that was used to locate each search result;

a click-through processor module operative to process the data to determine click-through behavior for each search result; and

a feedback processor operative to process the data to obtain explicit feedback on each search result.

16. The system of claim 14, wherein the session identifier module is further operative to identify one or more of the plurality of search results as a session of related search results by identifying one or more of the plurality of search results obtained in a prescribed time period.

17. The system of claim 14, wherein the relevance classifier module is further operative

to determine a first user behavior profile based on one or more relevance factors associated with a search result for which explicit feedback is provided and for which the search result is acceptable to a searching user;

to determine a second user behavior profile based on one or more relevance factors associated with a search result for which explicit feedback is provided and for which the search result is not acceptable to a searching user; and

to determine a third user behavior profile based on one or more relevance factors associated with a search result for which explicit feedback is provided and for which the search result is partially acceptable to a searching user.

18. The system of claim 17, wherein the relevance classifier module is further operative to apply the first, second and third user behavior profiles to one or more relevance factors calculated for each search result included in the session of search results for classifying each search result included in the session of search results for which no explicit feedback is provided as acceptable, unacceptable or partially acceptable.

19. A computer readable medium containing computer executable instructions which when executed perform a method of analyzing search behavior, comprising:

identifying one or more of the plurality of search results obtained in a prescribed time period as a session of related search results;

analyzing user behavior associated with a plurality of search results for which explicit feedback is provided for determining one or more relevance factors associated with search results that have been classified by a searching user as acceptable, unacceptable or partially acceptable;

calculating one or more relevance factors for each search result included in the session of search results;

comparing the calculated one or more relevance factors for each search result included in the session of search results to the determined one or more relevance factors associated with search results that have been classified by a searching user as acceptable, unacceptable or partially acceptable; and

providing a relevance classification of acceptable, unacceptable or partially acceptable for each search result included in the session of search results based on a relevance of each search result to a search query that was used to locate each search result and based on a similarity of the calculated one or more relevance factors for each search result included in the session of search results to the determined one or more relevance factors associated with search results that have been classified by a searching user as acceptable, unacceptable or partially acceptable.

20. The computer readable medium of claim 19, wherein analyzing user behavior associated with a plurality of search results for which explicit feedback is provided for determining one or more relevance factors associated with search results that have been classified by a searching user as acceptable, unacceptable or partially acceptable includes:

determining a first user behavior profile based on one or more relevance factors associated with a search result for which explicit feedback is provided and for which the search result is acceptable to a searching user;

determining a second user behavior profile based on one or more relevance factors associated with a search result for which explicit feedback is provided and for which the search result is not acceptable to a searching user; and

determining a third user behavior profile based on one or more relevance factors associated with a search result for which explicit feedback is provided and for which the search result is partially acceptable to a searching user.