Identifying and reporting unexpected behavior in targeted advertising environment

• US 20050144067A1
• Filed: 12/19/2003
• Published: 06/30/2005
• Est. Priority Date: 12/19/2003
• Status: Abandoned Application

• Current Assignee: Palo Alto Research Center, Inc. (Xerox Holdings Corp.)
• Original Assignee: Palo Alto Research Center, Inc. (Xerox Holdings Corp.)
• Inventors: Calabria, Hermann, Farahat, Ayman O., Greene, Daniel H., Chen, Francine R.
• Application Number:
  US10/742,380
• Publication Number:
  US 20050144067A1
• Time in Patent Office: Days
• US Class Current: 705/14.520
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• 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 30/0275 Auctions
• CPC Class Codes: G06Q 30/08 Auctions

Abstract

A method for generating or determining data sources useful for detecting non-conforming behavior associated with pay-per-click advertising in a keyword searching environment includes: a) observing behavior associated with the pay-per-click advertising, b) predicting behavior associated with the observed behavior, and c) comparing the observed behavior to the predicted behavior to identify unexpected behavior associated with the pay-per-click advertising. In another embodiment, a method of monitoring behavior associated with targeted advertising in a keyword searching environment is provided. In another aspect, an apparatus for monitoring behavior associated with targeted advertising in a keyword searching environment includes: at least one observed behavior model, at least one predicted behavior model, and at least one comparator logic process to identify non-confirming behavior associated with the targeted advertising.

Claims

1. A method for generating or determining data sources useful for detecting non-conforming behavior associated with pay-per-click advertising in a keyword searching environment, the method including the steps:

a) observing behavior associated with the pay-per-click advertising;

b) predicting behavior associated with the observed behavior; and

c) comparing the observed behavior to the predicted behavior to identify unexpected behavior associated with the pay-per-click advertising.

2. The method as set forth in claim 1 wherein the observed behavior is based on data from two or more components in the keyword searching environment.

3. The method as set forth in claim 2 wherein the data from the two or more components includes at least one of keyword data, advertisement content data, sponsored search results data, regular search results data, click-through data, advertiser web site content data, or regular search result web site content data.

4. The method as set forth in claim 2 wherein the two or more components include at least one of a keyword search engine, an advertiser web site, or a regular search result web site.

5. The method as set forth in claim 1 wherein the observed behavior includes at least one of click-throughs on sponsored search results by automated agents, low relevance sponsored advertisements in relation to corresponding keywords, low relevance advertiser web site in relation to corresponding keywords, or web spam in regular search result web site.

6. The method as set forth in claim 1, further including:

d) storing the unexpected behavior in a storage device.

7. The method as set forth in claim 6, further including:

e) reporting the unexpected behavior to an output device.

8. The method as set forth in claim 1 wherein the observed behavior is processed using at least one of an observed click-through behavior model, an observed keyword bid behavior model, an observed advertisement bid behavior model, an observed advertiser web site behavior model, an observed regular search result behavior model, an observed regular search result web site behavior model, or a topic analysis process.

9. The method as set forth in claim 1 wherein the predicted behavior is dynamically adjusted based on the observed behavior.

10. The method as set forth in claim 1 wherein the predicted behavior is processed using at least one of a predicted human user behavior model, a predicted automated agent behavior model, a predicted keyword, advertisement content, and advertiser web site content relevance behavior model, a predicted click-through behavior model, or a predicted regular search result web site behavior model.

20. The method as set forth in claim 1 further including:

providing the identified unexpected behavior to an output device including billing logic, used to determine a bill associated with the pay-per-click advertising;

providing the predicted behavior to the billing logic;

comparing the identified unexpected behavior with the predicted behavior; and

adjusting the bill associated with the pay-per-click advertising.

21. The method as set forth in claim 1 further including:

providing the identified unexpected behavior to an output device including billing logic, used to determine a bill associated with the pay-per-click advertising;

acknowledging, by the billing logic, that the identified unexpected behavior is over a determined threshold;

providing the predicted behavior to the billing logic;

providing the observed behavior to the billing logic;

comparing the predicted behavior and the observed behavior; and

adjusting the bill associated with the pay-per-click advertising based on the comparing.

11. A method of monitoring behavior associated with targeted advertising in a keyword searching environment, the method including the steps:

a) observing behavior associated with the targeted advertising;

b) predicting behavior associated with the observed behavior;

c) comparing the observed behavior to the predicted behavior to identify non-conforming behavior associated with the targeted advertising;

d) storing the non-conforming behavior on a storage device; and

e) reporting the non-conforming behavior to an output device.

12. The method as set forth in claim 11 wherein the observed behavior is observed using an advertisement including at least one valid image associated with a first hyper-link and at least one invalid image associated with a second hyper-link.

13. The method as set forth in claim 12 wherein the second hyper-link is not recognized as a pay-per-click event.

14. The method as set forth in claim 12 wherein one or more of the at least one valid image or the at least one invalid image is an image including text.

15. The method as set forth in claim 14 wherein the image has been modified using one or more methods for detecting an automated agent.

16. The method as set forth in claim 15 wherein the method for detecting the automated agent includes at least one of transformation or degradation.

17. The method as set forth in claim 11 wherein the observed behavior is processed using at least one of a latent semantic analysis model, a probabilistic latent semantic analysis model, a machine learning model, an information foraging model, a spreading activation model, or a Kalman filter.

18. The method as set forth in claim 11 wherein the predicted behavior is processed using at least one of a latent semantic analysis model, a probabilistic latent semantic analysis model, a machine learning model, an information foraging model, a spreading activation model, or a Kalman filter.

19. The method as set forth in claim 18 wherein the predicted behavior model uses at least one of an observed click-through behavior, an observed keyword bid behavior, an observed advertisement bid behavior, an observed advertiser web site behavior, an observed regular search result behavior, or an observed regular search result web site behavior.

22. An apparatus for monitoring behavior associated with targeted advertising in a keyword searching environment, the apparatus including:

at least one observed behavior model for identifying observed behavior associated with the targeted advertising;

at least one predicted behavior model for identifying predicted behavior associated with the observed behavior; and

at least one comparator logic process in communication with one or more of the at least one observed behavior model and one or more of the at least one predicted behavior model for comparing the observed behavior to the predicted behavior to identify non-conforming behavior associated with the targeted advertising.

23. The apparatus as set forth in claim 22, further including:

a storage device in communication with the each comparator logic process for storing the non-conforming behavior.

24. The apparatus as set forth in claim 23, further including:

an output device in communication with the storage device for reporting the non-conforming behavior to at least one of a user or one or more components of the keyword searching environment.

25. The apparatus as set forth in claim 22 wherein the at least one observed behavior model includes an observed click-through behavior model, wherein the at least one predicted behavior model includes a predicted automated agent behavior model.

26. The apparatus as set forth in claim 24, the at least one predicted behavior model further including:

a predicted human user behavior model.

27. The apparatus as set forth in claim 22 wherein the at least one observed behavior model includes an observed keyword bid behavior model, an observed advertisement bid behavior model, an observed advertiser web site behavior model, a topic analysis process, or an observed keyword bid, advertisement bid, and advertiser web site content relevance behavior model, wherein the at least one predicted behavior model includes a predicted keyword bid, advertisement bid, and advertiser web site content relevance behavior model, wherein the at least one comparator logic process includes a first comparator logic process in communication with the observed keyword bid, advertisement bid, and advertiser web site content relevance behavior model and predicted keyword bid, advertisement bid, and advertiser web site content relevance behavior model to identify non-conforming behavior associated with low relevance of any combination of keywords, advertisement content, and advertiser web site content.

28. The apparatus as set forth in claim 27 wherein the at least one observed behavior model includes an observed click-through behavior model, wherein the at least one predicted behavior model includes a predicted click-through behavior model in communication with the observed keyword bid, advertisement bid, and advertiser web site content relevance behavior model, wherein the at least one comparator logic process includes a second comparator logic process in communication with the observed click-through behavior model and predicted click-through behavior model to identify non-conforming behavior associated with click-through rates for advertisements in search results lists associated with the keywords.

29. The apparatus as set forth in claim 22 wherein the at least one observed behavior model includes an observed regular search result behavior model and an observed regular search result web site behavior model, wherein the at least one predicted behavior model includes a predicted regular search result web site behavior model behavior model.

30. A computer program product for use with an apparatus for monitoring behavior associated with targeted advertising in a keyword searching environment, the computer program product including:

a computer usable medium having computer readable program code embodied in the medium for causing;

i) observation of behavior associated with the targeted advertising;

ii) prediction of behavior associated with the observed behavior;

iii) comparison of the observed behavior to the predicted behavior to identify non-conforming behavior associated with the targeted advertising;

iv) storage of the non-conforming behavior on a storage device; and

v) reporting of the non-conforming behavior to an output device.