SYSTEM AND METHOD FOR MODELING USER SELECTION FEEDBACK IN A SEARCH RESULT PAGE

• US 20080256050A1
• Filed: 04/10/2007
• Published: 10/16/2008
• Est. Priority Date: 04/10/2007
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Sarukkai, Ramesh R., Zhang, Ruofei
• Granted Patent: US 7,827,184 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

Abstract

The present invention provides for improving the search relevance of a search results page by including a perceived relevance factor. The system, device and method monitors user selection of elements in the search results page, where these selections indicate relevance of the element compared with the original search request. A perceived relevance factor for the element is then determined based on probabilistic-based computations accounting for the element, which may include a description, a thumbnail and/or meta data, and the position of the element on the search results page. Thereby, for future searches and search results page generation, relevance factors may be calculated based on various factors, including the element attribute based relevant scores and the perceived relevance factor.

Claims

1. A method for improving search relevance for a user search, the method comprising:

generating a search results page in response to a search request, the search result page including document identifiers for each of a plurality of referenced documents retrieved in response to the search request;

monitoring user selection of one of the document identifiers;

determining a perceived relevance factor for the selected document identifier including prior probability calculations to determine relevance based on a position of the document identifier on the search results page and an inverse of the probability calculations to determine relevance of the identifier and identifier meta data; and

calculating a relevance factor for the selected document identifier based on the perceived relevance factor and a plurality of document attribute based relevant scores.

2. The method of claim 1 further comprising:

storing the relevance factor to a search index.

3. The method of claim 2 further comprising:

upon receipt of a new search request, accessing the search index to determine the relevance factor for the selected identifier; and

placing the selected identifier at a specific position in a new search results page based on the relevance factor.

4. The method of claim 1 wherein the user search relates to at least one of:

a video search and an image search.

5. The method of claim 1 wherein each identifier includes a thumbnail.

6. The method of claim 1 wherein the perceived relevance factor may be computed in an offline computation.

7. The method of claim 1 wherein the step of determining the perceived relevance factor based on the user selection includes modeling user selection activities with a statistical model.

8. The method of claim 7 wherein the statistical model is based on the probability calculations.

9. A system for improving search relevance for a user search, the system comprising:

a search database having a plurality of document identifiers for each of a plurality of referenced documents;

a processing device operative to, based on executable instructions, generate a search results page in response to a search request, the search results page including the plurality of document identifiers such that the search results page is presented to a user;

a user selection monitoring device operative to determine user selection of one of the document identifiers, and

the processing device further operative to, based on the executable instructions;

determine a perceived relevance factor for the selected document identifier including prior probability calculations to determine relevance based on a position of the document identifier on the search results page and an inverse of the probability calculations to determine relevance of the identifier and identifier meta data; and

calculate a relevance factor for the selected document identifier based on the perceived relevance factor and a plurality of document attribute based relevant scores.

10. The system of claim 9 further comprising:

a search index operatively coupled to the processing device, the search index having the relevance factor stored therein.

11. The system of claim 10 wherein the processing device, in response to the executable instruction, is further operative to:

upon receipt of a new search request, access the search index to determine the relevance factor to the selected identifier; and

place the selected identifier at a specific position in a new search results page based on the relevance factor.

12. The system of claim 9 wherein the user search relates to at least one of:

a video search and an image search.

13. The system of claim 9 wherein each identifier includes a thumbnail.

14. The system of claim 9 wherein the perceived relevance factor, as computed by the processing device in response to the executable instructions, may be computed in an offline computation.

15. The system of claim 9 wherein the executable operation of determining the perceived relevance factor based on the user selection as performed by the processing device includes modeling user selection activities with a statistical model.

16. The system of claim 15 wherein the statistical model is based on the probability calculations.

17. A computer readable medium having executable instructions stored thereon such that upon execution of the instructions, a processing device is operative to:

generate a search results page in response to a search request, the search result page including document identifiers for each of a plurality of referenced documents retrieved in response to the search request;

monitor user selection of one of the document identifiers;

determine a perceived relevance factor for the selected document identifier including prior probability calculations to determine relevance based on a position of the document identifier on the search results page and an inverse of the probability calculations to determine relevance of the identifier and identifier meta data; and

calculate a relevance factor for the selected document identifier based on the perceived relevance factor and a plurality of document attribute based relevant scores.

18. The computer readable medium of claim 17 having executable instructions stored thereon such that upon execution the processing device is further operative to:

store the relevance factor to a search index.

19. The computer readable medium of claim 18 having executable instructions stored thereon such that upon execution the processing device is further operative to:

upon receipt of a new search request, access the search index to determine the relevance factor for the selected identifier; and

place the selected identifier at a specific position in a new search results page based on the relevance factor.

20. The computer readable medium of claim 17 wherein the user search relates to at least one of:

a video search and an image search.

21. The computer readable medium of claim 17 wherein each identifier includes a thumbnail.

22. The computer readable medium of claim 17 wherein the perceived relevance factor may be computed in an offline computation.

23. The computer readable medium of claim 17 wherein the executable instruction of, as performed by the processing device, determining the perceived relevance factor based on the user selection includes modeling user selection activities with a statistical model.

24. The computer readable medium of claim 23 wherein the statistical model is based on the probability calculations.