Using estimated ad qualities for ad filtering, ranking and promotion

• US 20070156621A1
• Filed: 12/30/2005
• Published: 07/05/2007
• Est. Priority Date: 12/30/2005
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Pregibon, Daryl, Wright, Daniel, Tang, Diane
• Granted Patent: US 7,827,060 B2
• Time in Patent Office: Days
• US Class Current: 706/48
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 30/0243 Comparative campaigns
• CPC Class Codes: G06Q 30/0246 Traffic
• CPC Class Codes: G06Q 30/0251 Targeted advertisements
• CPC Class Codes: G06Q 30/0254 based on statistics
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 30/0271 Personalized advertisement

Abstract

A system obtains a first parameter (QP₁) associated with a quality of an advertisement among multiple advertisements, where the first quality parameter (QP₁) does not include a click through rate (CTR). The system functionally combines the first quality parameter (QP₁) with at least one other parameter and uses the functional combination to filter, rank or promote the advertisement among the multiple advertisements.

Claims

1. A method, comprising:

obtaining a first parameter (QP₁) associated with a quality of an advertisement among a plurality of advertisements, wherein the first quality parameter (QP₁) does not include a click through rate (CTR);

functionally combining the first quality parameter (QP₁) with at least one other parameter; and

using the functional combination to filter, rank or promote the advertisement among the plurality of advertisements.

2. The method of claim 1, wherein the at least one other parameter includes a click through rate (CTR).

3. The method of claim 1, wherein the first quality parameter (QP₁) comprises a probability that the advertisement is a good advertisement.

4. The method of claim 1, wherein the first quality parameter (QP₁) comprises a probability that the advertisement is a bad advertisement.

5. The method of claim 1, wherein filtering the advertisement comprises preventing the advertisement from being provided to a user.

6. The method of claim 2, wherein functionally combining the first quality parameter (QP₁) with at least one other parameter comprises:

multiplying the first quality parameter (QP₁) with the click through rate (CTR);

QP₁*CTR.

7. The method of claim 6, wherein using the function combination to filter the advertisement comprises:

comparing QP₁*CTR with a threshold to determine whether to filter the advertisement.

8. The method of claim 6, wherein using the functional combination to promote the advertisement among the plurality of advertisements comprises:

comparing QP₁*CTR with a threshold to determine whether to promote the advertisement among the plurality of advertisements.

9. The method of claim 2, wherein the at least one other parameter further includes a cost-per-click (CPC) associated with the advertisement.

10. The method of claim 9, wherein functionally combining the first quality parameter (QP₁) with the at least one other parameter comprises:

multiplying the first quality parameter (QP₁) by CTR and CPC;

QP₁*CPC*CTR.

11. The method of claim 10, wherein using the functional combination to rank the advertisement comprises:

ranking the advertisement among the plurality of advertisements using QP*CPC*CTR.

12. The method of claim 9, further comprising:

obtaining a second parameter (QP₂) associated with a quality of the advertisement, wherein the second quality parameter (QP₂) does not include the click through rate (CTR); and

functionally combining the second quality parameter (QP₂) with the first quality parameter (QP₁) and the at least one other parameter.

13. The method of claim 12, wherein functionally combining the second quality parameter (QP₂) with the first quality parameter (QP₁) and the at least one other parameter comprises:

CTR*CPC+K₁*QP₁*CTR−

K₂*QP₂*CTR, where K₁ and K₂ are constants.

14. The method of claim 13, wherein using the functional combination to promote the advertisement among the plurality of advertisements comprises:

comparing CTR*CPC+K₁*QP₁*CTR−

K₂*QP₂*CTR with a threshold to determine whether to promote the advertisement among the plurality of advertisements.

16. The method of claim 1, wherein deriving the statistical model comprises:

using at least one of logistic regression, regression trees or boosted stumps to generate the statistical model.

17. The method of claim 1, further comprising:

ranking advertisements of the subset of advertisements, based on the estimated quality scores, to order the subset of advertisements.

18. The method of claim 17, further comprising:

providing the subset of advertisements in the ranked order to the users.

19. The method of claim 17, further comprising:

promoting one or more of the subset of advertisements based on the estimated quality scores.

20. The method of claim 19, further comprising:

positioning the promoted one or more advertisements of the subset of advertisements in a prominent position on a document; and

positioning unpromoted advertisements of the second group of advertisements in a less prominent position on the document than the promoted one or more advertisements.

21. The method of claim 20, wherein providing the subset of advertisements to a user comprises:

providing the document to the user.

15. A method, comprising:

obtaining ratings associated with a first group of advertisements, where the ratings indicate a quality of the first group of advertisements;

observing multiple different user actions associated with user selection of advertisements of the first group of advertisements;

deriving a statistical model using the observed user actions and the obtained ratings;

using the statistical model to estimate quality scores associated with a second group of advertisements; and

providing a subset of advertisements of the second group of advertisements to a user based on the estimated quality scores.

22. A method, comprising:

determining quality scores associated with a set of advertisements using a statistical model, where the quality scores do not include a click through rate (CTR);

disabling a first subset of advertisements of the set of advertisements based on the determined quality scores; and

providing a second subset of the set of advertisements to a user, where the second subset of the set of advertisements comprises the set of advertisements minus the first subset of advertisements.

23. The method of claim 22, further comprising:

ranking advertisements of the second subset of the set of advertisements based on the determined quality scores.

24. The method of claim 23, further comprising:

promoting one or more of the second subset of advertisements based on the determined quality scores.

25. The method of claim 24, further comprising:

positioning the promoted one or more advertisements of the second subset of advertisements in a first position on a document; and

positioning unpromoted advertisements of the set of advertisements in a second position on the document that is different than the first position.

26. The method of claim 25, wherein providing the second subset of advertisements to a user comprises:

providing the document to the user.

27. A method, comprising:

determining quality scores associated with a set of advertisements using a statistical model, where the quality scores do not include a click through rate (CTR); and

ranking advertisements of the set of advertisements based on the determined quality scores to determine a ranked order.

28. The method of claim 27, further comprising:

providing the ranked advertisements to a user in the ranked order.

29. A method, comprising:

determining quality scores associated with a group of advertisements using a statistical model where the quality scores do not include a click through rate (CTR);

promoting one or more advertisements of the group of advertisements based on the determined quality scores;

positioning the promoted one or more advertisements of the group of advertisements in a prominent position on a document; and

positioning unpromoted advertisements of the group of advertisements in a less prominent position on the document than the promoted one or more advertisements.

30. A computer-readable medium that stores computer-executable instructions, comprising:

instructions for estimating quality scores associated with a set of advertisements using a statistical model where quality scores do not include a click through rate (CTR);

instructions for disabling a first subset of advertisements of the set of advertisements based on the estimated quality scores;

instructions for ranking advertisements of a second subset of advertisements, based on the estimated quality scores, to order the second subset of advertisements, where the second subset of advertisements comprises the set of advertisements minus the first subset of advertisements;

instructions for promoting one or more of the ranked second subset of advertisements based on the estimated quality scores; and

instructions for positioning the promoted one or more advertisements of the ranked second subset of advertisements in a first position on a document and positioning unpromoted advertisements of the ranked subset of advertisements in a second position on the document that is different than the first position.

31. A system, comprising:

means for determining quality scores associated with a group of advertisements using a statistical model, where quality scores do not include a click through rate (CTR);

means for disabling a first subset of advertisements of the first group of advertisements based on the determined quality scores;

means for ranking advertisements of a second subset of advertisements of the group of advertisements, based on the determined quality scores, to order the second subset of advertisements;

means for promoting one or more of the ranked second subset of advertisements based on the estimated quality scores; and

means for positioning the promoted one or more advertisements of the ranked second subset of advertisements in a more prominent position on a document than unpromoted advertisements of the ranked second subset of advertisements.