System & method for influencing recommender system

• US 20040267604A1
• Filed: 06/07/2004
• Published: 12/30/2004
• Est. Priority Date: 06/05/2003
• Status: Active Grant

• Current Assignee: Xylon LLC (Intellectual Ventures LLC)
• Original Assignee: Hayley Logistics LLC (Intellectual Ventures LLC)
• Inventors: Gross, John N.
• Granted Patent: US 8,103,540 B2
• Time in Patent Office: Days
• US Class Current: 705/7.29
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 30/0203 Market surveys; Market polls
• CPC Class Codes: G06Q 30/0204 Market segmentation

Abstract

A system and method for implementing/influencing a recommender system which provides recommendations to users based on characteristics of certain trendsetters within a member population. The trendsetters are determined by studying historical adoption behavior of a group within the member population, or by reference to known indicia.

Claims

1. A method of influencing behavior of an electronic recommender system, which recommender is used for suggesting items of interest to a community of online subscribers, the method comprising the steps of:

(a) identifying trendsetters, said trendsetters being individuals who are characterized as relatively early adopters of items that later become relatively popular within the community of subscribers;

(b) measuring an adoption rate by said trendsetters for a particular item;

(c) modifying recommendations provided by the recommender system for items, including said particular item, based on a value of said adoption rate.

2. The method of claim 1 wherein the items include human readable content, products and/or services.

3. The method of claim 1, wherein a trendsetter is selected by identifying an item that achieves a greater than 50% adoption rate within the community of subscribers, and then identifying the first N persons who adopted such item.

4. The method of claim 1, wherein said trendsetters are users, purchasers, and/or renters of said items.

5. The method of claim 1, wherein the recommender system uses a collaborative filtering algorithm, and said trendsetters are weighted more heavily than other subscribers within the community of online subscribers.

6. The method of claim 1, wherein at least some of said trendsetters are not subscribers in the community of online subscribers.

7. The method of claim 1, further including a step of:

identifying trend rejecters for said particular item, and modifying a recommender system recommendation by reducing contributions by said trend rejecters for said particular item.

8. The method of claim 1, further including a step of:

identifying particular items rejected by said trendsetters, and modifying a recommendation for said particular items based on a rejection rate exhibited by said trendsetters.

9. A method of influencing behavior of an electronic recommender system, which recommender is used for suggesting items of interest to a community of online subscribers, the method comprising the steps of:

(a) identifying trendsetters, said trendsetters being individuals who are characterized as relatively early adopters of items that later become relatively popular within the community of subscribers;

(b) measuring a trendsetter rating for a particular item provided by said trendsetters;

(c) modifying a user rating for said particular item for other subscribers based on said trendsetter rating;

wherein said trendsetter rating is used as part of generating recommendations for items to a user requesting a suggestion from the electronic recommender system.

10. The method of claim 9, further including a step of:

identifying particular items rejected by said trendsetters, and generating a recommendation for said particular items based on a rejection rate exhibited by said trendsetters.

11. The method of claim 9, wherein said trendsetters are associated with web pages on the Internet.

12. A system for providing recommendations of items of interest to a community of online subscribers, the system comprising:

(a) a software module which is adapted for;

i. identifying trendsetters, said trendsetters being individuals who are characterized as relatively early adopters of items that later become relatively popular within the community of subscribers;

ii. measuring an adoption rate by said trendsetters for a particular item;

iii. provided recommendations for items, including said particular item, based on a value of said adoption rate.

13. The system of claim 12, wherein said trendsetters are users, purchasers, and/or renters of said items.

14. The system of claim 12, wherein said software module includes a collaborative filtering algorithm in which said trendsetters are weighted more heavily than other subscribers within the community of online subscribers.

15. The system of claim 12, wherein at least some of said trendsetters are not subscribers in the community of online subscribers.

16. The system of claim 12, wherein the software module is further adapted for identifying trend rejecters for said particular item, and providing a recommendation by reducing contributions by said trend rejecters for said particular item.

17. The system of claim 12, further including a software module for adjusting advertising presented with said recommendations based on adoption rate.

18. The system of claim 12, wherein the trendsetters are associated with individual web pages.