Modifying search result ranking based on implicit user feedback

• US 8,661,029 B1
• Filed: 11/02/2006
• Issued: 02/25/2014
• Est. Priority Date: 11/02/2006
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Kim, Hyung-Jin, Tong, Simon, Shazeer, Noam M., Diligenti, Michelangelo
• Primary Examiner(s): LU, KUEN S
• Application Number:
  US11/556,143
• Time in Patent Office: 2,672 Days
• Field of Search: None
• US Class Current: 707/723
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Systems and techniques relating to ranking search results of a search query include, in general, subject matter that can be embodied in a computer-implemented method that includes determining a measure of relevance for a document result within a context of a search query for which the document result is returned, the determining being based on a first number in relation to a second number, the first number corresponding to longer views of the document result, and the second number corresponding to at least shorter views of the document result; and outputting the measure of relevance to a ranking engine for ranking of search results, including the document result, for a new search corresponding to the search query. The subject matter described in this specification can also be embodied in various corresponding computer program products, apparatus and systems.

Claims

1. A computer-implemented method comprising:

obtaining a measure of relevance for a document referenced by a search result wherein the search result was provided in response to a search query,

the measure of relevance being based on a first number in relation to a second number, the first number corresponding to a subset of a count of views of the document following selections of the search result when the search result was provided in response to the search query, and the second number corresponding to the count of views of the document following selections of the search result; and

providing the measure of relevance to a ranking engine for ranking of search results, for the document referenced by the search result.

2. The method of claim 1 wherein the measure of relevance is calculated by dividing the first number by the second number.

3. The method of claim 1 wherein the first number is a weighted combination of a plurality of the views of the document wherein one or more of the views is weighted using a respective weight based at least in part on a respective viewing duration of the document.

4. The method of claim 1 wherein the first number is a total number of long clicks on the search result when the search result was provided in response to the search query, and wherein the second number is a total number of clicks on the search result when the search result was provided in response to the search query.

5. The method of claim 3 wherein the weight is determined using a continuous function of viewing length.

6. The method of claim 3 wherein the weight is determined using a discontinuous function of viewing length.

7. The method of claim 6, further comprising:

classifying the viewing length into a viewing time category of a plurality of viewing time categories; and

determining the weight based at least in part on the viewing time category.

8. The method of claim 3 wherein the weight is further based on using a viewing length differentiator.

9. The method of claim 8, wherein the viewing length differentiator includes a factor governed by a determined category of the search query, and the weight is further based at least in part on the determined category of the search query.

10. The method of claim 8, wherein the viewing length differentiator includes a factor governed by a determined type of a user generating the individual selections, and the weight is further based at least in part on the determined type of the user.

11. A computer-readable medium storing a computer program product that, when executed by data processing apparatus, causes the data processing apparatus to perform operations comprising:

obtaining a measure of relevance for a document referenced by a search result wherein the search result was provided in response to a search query, the measure of relevance being based on a first number in relation to a second number, the first number corresponding to a subset of a count of views of the document following selections of the search result when the search result was provided in response to the search query, and the second number corresponding to the count of views of the document following selections of the search result; and

providing the measure of relevance to a ranking engine for ranking of search results, for the document referenced by the search result.

12. The computer-readable medium of claim 11 wherein the measure of relevance is calculated by dividing the first number by the second number.

13. The computer-readable medium of claim 11 wherein the first number is a weighted combination of a plurality of the views of the document wherein one or more of the views is weighted using a respective weight based at least in part on a respective viewing duration of the document.

14. The computer-readable medium of claim 11, wherein the first number is a total number of long clicks on the search result when the search result was provided in response to the search query, and wherein the second number is a total number of clicks on the search result when the search result was provided in response to the search query.

15. The computer-readable medium of claim 13 wherein the weight is determined using a continuous function of viewing length.

16. The computer-readable medium of claim 13 wherein the weight is determined using a discontinuous function of viewing length.

17. The computer-readable medium of claim 16, wherein the operations further comprise:

classifying the viewing length into a viewing time category of a plurality of viewing time categories; and

determining the weight based at least in part on the viewing time category.

18. The computer-readable medium of claim 13 wherein the weight is further based on using a viewing length differentiator.

19. The computer-readable medium of claim 18, wherein the viewing length differentiator includes a factor governed by a determined category of the search query, and the weight is further based at least in part on the determined category of the search query.

20. The computer-readable medium of claim 18, wherein the viewing length differentiator includes a factor governed by a determined type of a user generating the individual selections, and the weight is further based at least in part on the determined type of the user.

21. A system comprising:

data processing apparatus programmed to perform operations comprising;

obtaining a measure of relevance for a document referenced by a search result wherein the search result was provided in response to a search query, the measure of relevance being based on a first number in relation to a second number, the first number corresponding to a subset of a count of views of the document following selections of the search result when the search result was provided in response to the search query, and the second number corresponding to the count of views of the document following selections of the search result; and

providing the measure of relevance to a ranking engine for ranking of search results, for the document referenced by the search result.

22. The system of claim 21 wherein the measure of relevance is calculated by dividing the first number by the second number.

23. The system of claim 21 wherein the first number is a weighted combination of a plurality of the views of the document wherein one or more of the views is weighted using a respective weight based at least in part on a respective viewing duration of the document.

24. The system of claim 23 wherein the weight is determined using a continuous function of viewing length.

25. The system of claim 23 wherein the weight is determined using a discontinuous function of viewing length.

26. The system of claim 23 wherein the weight is further based on using a viewing length differentiator.

27. The system of claim 26, wherein the viewing length differentiator includes a factor governed by a determined category of the search query, and the weight is further based at least in part on the determined category of the search query.

28. The system of claim 26, wherein the viewing length differentiator includes a factor governed by a determined type of a user generating the individual selections, and the weight is further based at least in part on the determined type of the user.

29. The system of claim 21, wherein the first number is a total number of long clicks on the search result when the search result was provided in response to the search query, and wherein the second number is a total number of clicks on the search result when the search result was provided in response to the search query.

30. The system of claim 29, wherein the operations further comprise:

classifying the viewing length into a viewing time category of a plurality of viewing time categories; and

determining the weight based at least in part on the viewing time category.