System and method for influencing recommender system & advertising based on programmed policies

• US 7,689,432 B2
• Filed: 06/07/2004
• Issued: 03/30/2010
• Est. Priority Date: 06/06/2003
• Status: Active Grant

• Current Assignee: Xylon LLC (Intellectual Ventures LLC)
• Original Assignee: Hayley Logistics LLC (Intellectual Ventures LLC)
• Inventors: Gross, John N.
• Primary Examiner(s): OUELLETTE, JONATHAN P
• Application Number:
  US10/863,743
• Publication Number:
  US 20040260574A1
• Time in Patent Office: 2,122 Days
• Field of Search: 705/1
• US Class Current: 705/1.1
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A system and method for influencing a recommender system and advertising uses a controlled, programmed policy. The recommender system uses the policy to adjust recommendations made to users for particular items, so that a policy holder receives some measure of preference in recommendations. The preference may be implemented in the form of a filter, or a presentation of items. The policy can be applied selectively, and be based on pay for placement type consideration, demographics, time, and other related parameters. Advertising for the users can be similarly adjusted in coordination with the policy.

Claims

1. A method of operating an online recommender system for recommending items of interest to users interacting with an online service provider, comprising:

defining a pay-for-placement influence policy applicable to a third party advertiser;

establishing a biasing policy based on the influence policy for influencing recommendations made by the recommender system to a user of the recommender system,

wherein the biasing policy is representative of preferences of the third party advertiser, and biases the recommender system to recommend items associated with the third party advertiser;

responding to a request for a recommendation; and

applying the biasing policy to the recommender system when presenting a policy influenced recommendation in response to the request.

2. The method of claim 1, further comprising prior to said applying:

evaluating a plurality of biasing policies under a cost-benefit analysis to determine an appropriate policy to apply,

wherein cost in the cost-benefit analysis is an error cost, the error cost being a difference in ranking of a non-biased proposed recommendation and a pay-for-placement policy influenced recommendation, and a benefit is a cost charged to the third party advertiser for the policy influenced recommendation.

3. The method of claim 2, further comprising:

controlling advertising material presented to the user with the policy influenced recommendation based on which of the plurality of biasing policies is used to generate the policy influenced recommendation.

4. The method of claim 2, wherein a different biasing policy can be used for each request for a recommendation made by the user, so that the plurality of biasing policies are utilized on a request by request basis.

5. The method of claim 1, further comprising:

comparing a policy influenced recommendation to the user against a recommendation made by the recommender system in the absence of considering the biasing policy.

6. The method of claim 2, further comprising:

updating a policy record for a selected biasing policy within a policy log when a policy influenced recommendation is made to the user based on said selected biasing policy.

7. The method of claim 6, further comprising:

generating a report of a number of policy influenced recommendations made by the recommender system broken down according to a type of biasing policy used for the policy influenced recommendations.

8. The method of claim 1, wherein the preferences comprise at least one of a preference filter to be used against a list of recommendation items generated by the recommender system, and a presentation preference for a plurality of items associated with the third party advertiser and which are part of the list of recommendation items.

9. The method of claim 1, wherein the biasing policy is defined with reference to a period of time in which the biasing policy is to be used.

10. The method of claim 1, wherein the biasing policy is defined with reference to a keyword determining if and when the biasing policy is to be used.

11. The method of claim 1, wherein multiple biasing policies are evaluated for each policy influenced recommendation made by the recommender system.

12. The method of claim 1, wherein the recommender system is implemented at an e-commerce website.

13. The method of claim 1, wherein the recommender system is implemented at least in part in a client computing device, including a programming receiver.

14. The method of claim 1, wherein the biasing policy is considered after a recommender system has generated an initial list of uninfluenced recommendations.

15. The method of claim 1, wherein the biasing policy is utilized by a recommender system while formulating an initial list of recommendations.

16. The method of claim 1, wherein the biasing policy includes a fulfillment requirement, so that the recommender system is forced to generate a number of pay-for-placement policy influenced recommendations based on the biasing policy.

17. The method of claim 2, wherein the biasing policies are selected on a controlled opportunity basis to influence recommendations, such that a first policy can have a higher percentage chance of influencing a recommendation compared to a second policy.

18. A method of presenting advertising via an electronic device in conjunction with recommendations made by a recommender system, comprising:

defining a pay-for-placement influence policy applicable to a third party advertiser;

implementing a biasing policy based on the influence policy for influencing recommendations made by the recommender system to a user of the recommender system,

wherein said biasing policy is representative of preferences of the third party advertiser, and biases the recommender system to recommend items associated with the third party advertiser;

responding to a request for a recommendation; and

presenting advertising content based on the biasing policy while presenting policy influenced recommendation to the user in response to the request.

19. The method of claim 18, wherein the request is made by the user during an online Internet session.

20. The method of claim 18, wherein the request is made by the user from a home based content receiver.

21. The method of claim 18, wherein advertising content is based on recommendation content presented in the policy influenced recommendation.

22. The method of claim 21, wherein the third party advertiser associated with policy influenced recommendation is automatically selected as a source of the advertising for the advertising content.

23. The method of claim 18, wherein the advertising content is presented based on a pay for placement request from the third party advertiser.

24. A recommender system operating on a website and programmed to provide recommendations to users for items in response to a request for a recommendation, comprising:

a first software routine for defining a pay-for-placement influence policy applicable to a third party advertiser;

a second software routine for establishing a biasing policy based on the influence policy for influencing recommendations made by the recommender system to a user of the recommender system,

wherein the biasing policy is representative of preferences of the third party advertiser, and biases the recommender system to recommend items associated with the third party advertiser, and

wherein the recommender system applies the biasing policy to present policy influenced recommendations to the user in response to the request for a recommendation.

25. A method of recommending items of interest to a user of an online service provider, comprising:

maintaining a profile for a user at the online service provider, wherein the profile includes prior selections and/or ratings of items made by the user;

providing one or more recommendations to the user using a recommender system which uses account profiles of users similar to the user to generate such recommendations;

defining a pay-for-placement influence policy applicable to a third party advertiser;

defining a first item which is to receive a preference from the recommender system based on the influence policy,

wherein the preference biases the recommender system to recommend items associated with the third party advertiser; and

selectively influencing the recommender system on a subscriber by subscriber basis, so that the first item is presented to a user in a policy influenced recommendation based on the preference when the user'"'"'s profile meets criteria of the influence policy,

wherein the influence policy can be used to present recommendations of items to the user which would not otherwise be presented by the recommender system to the user.

26. The method of claim 25, wherein the policy influenced recommendation is presented during an interactive online session.

27. The method of claim 26, wherein the policy influenced recommendation is positioned and/or displayed more prominently than other items presented in response to a recommendation request.

28. The method of claim 25, wherein the policy influenced recommendation is provided based on the first item being a new item.

29. The method of claim 25, wherein the policy influenced recommendation is provided only during a certain predefined time period and/or during certain times of the day.

30. The method of claim 25, wherein the criteria includes demographic characteristics which must be associated with the profile for the user.

31. A method of recommending items of interest to a user of an online service provider, comprising:

maintaining a profile for a user at the online service provider, wherein the profile includes prior selections and/or ratings of items made by the user;

providing recommendations in a first list to the user over an electronic communications network using a recommender system which uses account profiles of other users similar to the user to generate such recommendations;

defining a pay-for-placement influence policy applicable to a third party advertiser;

defining a first item which is to be recommended by the recommender system based on the influence policy and preference factors unrelated to the account profiles of other users,

wherein the preference factors bias the recommender system to recommend items associated with the third party advertiser; and

selectively influencing the recommender system such that the first item is presented, based on the influence policy and the preference factors, only to selected users residing within predefined geographic regions,

wherein the influence policy can be used to present recommendations of items to the user which would not otherwise be presented by the recommender system over the electronic communications network to the user in the first list.

32. The method of claim 31, wherein the first item is a newly released movie.

33. The method of claim 31, wherein the first item is recommended in response to keywords used in requests for recommendations by users in the predefined geographic regions.

34. The method of claim 31, further comprising:

tabulating user selections of the first item made in response to the preference factors.

35. The method of claim 31, wherein the preference factors include an age of the user, such that the first item is presented only to selected users residing within predefined geographic regions who satisfy an age range criterion.

36. The method of claim 31, further comprising:

defining a second item which is to be recommended by the recommender system based on the influence policy and second preference factors unrelated to the account profiles of other users;

defining a schedule to be used in connection with recommending the first item and the second item, the schedule being based on a cost-benefit analysis performed on a subscriber by subscriber basis; and

presenting one or both of the first item and the second item to the selected users residing within the predefined geographic regions based on the schedule.

37. A method of recommending items of interest to users of an online service provider, comprising:

defining a pay-for-placement influence policy applicable to a third party advertiser;

providing an influenced recommendation based on a biasing policy derived from the influence policy for a first item to a user over a network using a collaborative filtering based recommendation engine,

wherein the biasing policy is representative of preferences of the third party advertiser, and biases the recommender system to recommend items associated with the third party advertiser, and

wherein the first item is characterized by a first content associated with a first identifiable source of content; and

providing an advertisement to the user which is correlated to the influenced recommendation, the advertisement being generated by an advertising delivery system and describing a second item to the user over the network,

wherein the second item is characterized by a second content that is also associated with the first identifiable source of content, and

wherein presentation of recommendations and advertisements are coordinated for users the user of the online service provider.

38. The method of claim 37, wherein the recommendation and the advertisement are presented at the same time.

39. The method of claim 38, wherein the recommendation and the advertisement are presented within a single web page.

40. A method of recommending items of interest to users of an online service provider, comprising:

defining a pay-for-placement influence policy applicable to a third party advertiser;

providing an influenced recommendation based a biasing policy derived from the influence policy for a first item to a user over a network using a collaborative filtering based recommendation engine,

wherein the biasing policy is representative of preferences of the third party advertiser, and biases the recommender system to recommend items associated with the third party advertiser; and

providing an advertisement to the user which is correlated to the recommendation, the advertisement being generated by an advertising delivery system and describing a second item to the user over the network,

wherein the second item is related logically to the first item and is associated with a context presented the user in which the first item is presented,

wherein presentation of recommendations and advertisements are coordinated for the user of the online service provider.

41. The method of claim 40, wherein the recommendation and the advertisement are presented at the same time.

42. The method of claim 41, wherein the recommendation and the advertisement are presented within a single web page.

43. A recommender system for recommending items of interest to users interacting with an online service provider, comprising:

means for defining a pay-for-placement influence policy applicable to a third party advertiser;

means for providing a biasing policy based on the influence policy to influence recommendations made by the recommendation system,

wherein the biasing policy is representative of preferences of the third party advertiser, and biases the recommender system to recommend items associated with the third party advertiser;

means for responding to a request for a recommendation from a user; and

means for implementing the biasing policy when presenting recommendations to the user in response to the request.

44. The recommender system of claim 43, further comprising:

means for presenting advertising content to the user based on the biasing policy when presenting recommendations to the user in response to the request.

45. A computer-readable medium having stored thereon, computer-executable instructions that, if executed by a computing device, cause the computing device to perform a method comprising:

defining a pay-for-placement influence policy applicable to a third party advertiser;

establishing a biasing policy based on the influence policy for influencing recommendations made by the recommender system to a user of the recommender system,

wherein the biasing policy is representative of preferences of the third party advertiser, and biases the recommender system to recommend items associated with the third party advertiser;

responding to a request for a recommendation; and

applying the biasing policy to the recommender system when presenting a policy influenced recommendation in response to the request.

46. A method for influencing a recommender system, comprising:

defining a pay-for-placement influence policy applicable to a third party advertiser;

specifying third party advertiser preferences from the influence policy;

implementing the third party advertiser preferences to bias the recommender system to items associated with the third party advertiser;

responding to a recommendation request from a user;

generating a policy influenced recommendation for the user by enforcing the influence policy;

presenting the policy influenced recommendation to the user; and

evaluating performance of the influence policy.