System and process for improving recommendations for use in providing personalized advertisements to retail customers

• US 8,583,524 B2
• Filed: 05/06/2008
• Issued: 11/12/2013
• Est. Priority Date: 05/06/2008
• Status: Active Grant

• Current Assignee: RichRelevance Incorporated
• Original Assignee: RichRelevance Incorporated
• Inventors: Selinger, David, Kohn, Tyler, DeCourcey, Michael, Ahuja, Sundeep, Osial, James, Sunwoo, Albert
• Primary Examiner(s): Kazimi, Hani M
• Assistant Examiner(s): Ali, Hatem M
• Application Number:
  US12/151,374
• Publication Number:
  US 20090281895A1
• Time in Patent Office: 2,016 Days
• Field of Search: 705/14
• US Class Current: 705/35
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0244 Optimization
• CPC Class Codes: G06Q 30/0246 Traffic
• CPC Class Codes: G06Q 30/0251 Targeted advertisements

Abstract

A method of improving user set recommendations for product advertising including receiving a request for user set recommendations from any of a set of retailers where such request is related to one or more products, receiving from a plurality of user sets from one or more automated user recommendation systems, wherein the plurality of user sets are generated using different user models and using ensemble learning to select one or more most relevant user sets from the plurality of user sets.

Claims

1. A computer-implemented method comprising:

receiving from a retailer, by a configured computer system, a request for one or more recommended user sets that include indications of multiple specific users to which to provide product advertising for one or more products, wherein the request indicates the one or more products but does not include the one or more user sets;

obtaining, by the configured computer system, a plurality of user sets from one or more automated user recommendation systems, the plurality of user sets having been generated using multiple different user models; and

in response to the receiving of the request,

using ensemble learning to automatically select, by the configured computer system and from the plurality of user sets, one or more most relevant user sets for the one or more products, the selected one or more user sets indicating multiple users, wherein the using of the ensemble learning includes automatically selecting the one or more most relevant user sets from the plurality of user sets based, at least in part, on information regarding user behavior collected from user activity subsequent to previous recommendations; and

electronically communicating the selected one or more user sets to the retailer, to enable the retailer to provide advertising for the one or more products to the multiple users of the selected one or more user sets, the electronic communicating of the selected one or more user sets being performed by the configured computer system.

2. The method of claim 1 wherein the automatic selecting includes using a naive Bayesian network to automatically select the one or more most relevant user sets from the plurality of user sets.

3. The method of claim 2 wherein the naive Bayesian network utilizes information in the automatic selecting concerning the retailer from which the request was received, a Web page for the retailer on which a recommendation is to be displayed, and the one or more products.

4. The method of claim 3 wherein the automatic selecting further includes using an observed strength and reliability value for previous selection of user sets in combination with a strength and reliability value predicted from the naive Bayesian network.

5. The method of claim 1 further comprising applying one or more filters to one or more of the plurality of user sets.

6. The method of claim 5 further comprising using ensemble learning to select the one or more filters.

7. The method of claim 1 further comprising, after the electronic communicating of the selected one or more user sets, providing the multiple users of the one or more user sets with product advertising for the one or more products on a retail website of the retailer.

8. The method of claim 1 further comprising, after the electronic communicating of the selected one or more user sets, providing the multiple users of the one or more user sets with product advertising for the one or more products without using a retail website.

9. A non-transitory computer-readable storage medium having computer-executable instructions stored thereon that when executed configure a computer system to perform the steps of:

receiving from a retailer a request for one or more recommended user sets that include indications of multiple specific users to which to provide product advertising, wherein the request is related to one or more products but does not include the one or more user sets, the receiving of the request being performed by the configured computer system;

receiving a plurality of user sets from one or more automated user recommendation systems, the plurality of user sets having been generated using multiple different user models, the receiving of the plurality of user sets being performed by the configured computer system; and

in response to the receiving of the request,

automatically selecting, by the configured computer system, one or more most relevant user sets from the plurality of user sets, the selected one or more user sets indicating multiple users, the automatic selecting being based at least in part on information regarding user activity subsequent to previous recommendations; and

electronically communicating the selected one or more user sets to the retailer, to enable the retailer to provide advertising for the one or more products to the multiple users of the selected one or more user sets, the electronic communicating of the selected one or more user sets being performed by the configured computer system.

10. The non-transitory computer-readable medium of claim 9 wherein the automatic selecting of the one or more most relevant user sets from the plurality of user sets is based at least in part on using a naive Bayesian network.

11. The non-transitory computer-readable medium of claim 10 wherein the naive Bayesian network utilizes information in the automatic selecting concerning the retailer from which the request was received, a page on which a recommendation is to be displayed, and the one or more products.

12. The non-transitory computer-readable medium of claim 10 wherein the automatic selecting of the one or more most relevant user sets from the plurality of user sets includes using ensemble learning based on an observed strength and reliability value for previous selection of user sets in combination with a strength and reliability value predicted from the naive Bayesian network.

13. A system comprising:

a memory; and

a controller associated with the memory, wherein the controller is configured to;

receive from a retailer a request for one or more user sets that include indications of multiple specific users to which to provide product advertising, wherein the request is related to one or more products but does not include the one or more user sets;

receive a plurality of user sets from one or more automated user recommendation systems, the plurality of user sets having been generated using multiple different user models;

in response to the received request, use ensemble learning to select one or more most relevant user sets from the plurality of user sets for use in responding to the received request, wherein use of the ensemble learning includes automatically selecting the one or more most relevant user sets from the plurality of user sets based, at least in part, on information regarding user behavior subsequent to previous recommendations; and

electronically communicate the selected one or more user sets to the retailer, to enable the retailer to provide advertising for the one or more products to multiple users of the selected one or more user sets.

14. The system of claim 13 wherein the using of the ensemble learning further includes automatically selecting the one or more most relevant user sets from the plurality of user sets based, at least in part, on using a naive Bayesian network.

15. The system of claim 14 wherein the naive Bayesian network utilizes information in the automatic selecting concerning the retailer from which the request was received, a page on which a recommendation is to be displayed, and the one or more products.

16. The system of claim 15 wherein the using of the ensemble learning further includes using an observed strength and reliability value for previous selection of user sets in combination with a strength and reliability value predicted from the naive Bayesian network.