Serving advertisements using entertainment ratings in a collaborative-filtering system

• US 20080147500A1
• Filed: 12/15/2006
• Published: 06/19/2008
• Est. Priority Date: 12/15/2006
• Status: Abandoned Application

• Current Assignee: Oath Inc. (Verizon Communications Inc.)
• Original Assignee: Oath Inc. (Verizon Communications Inc.)
• Inventors: Marlin, Ben, Slaney, Malcolm
• Application Number:
  US11/640,160
• Publication Number:
  US 20080147500A1
• Time in Patent Office: Days
• US Class Current: 705/14
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: G06Q 30/0263 based upon Internet or website rating

Abstract

A collaborative-filtering based combination model is used to select advertisements to serve to a current network user. The combination model comprises a plurality of vectors, each vector comprising ratings for entertainment items and ratings for advertisements (or advertisement categories) from a previous network user. A vector may also include ratings for demographic information of the previous network user. Upon receiving one or more entertainment item ratings (and optionally demographic ratings) of a current user, the combination model is used to select one or more advertisements to serve to the current user. A representative vector in the combination model is determined having entertainment item ratings (and optionally demographic ratings) that are most similar to the corresponding entertainment item ratings (and optionally demographic ratings) of the current user. The advertisement/advertisement category ratings of the representative vector are examined to select the advertisements to serve to the current user.

Claims

1. A method for implementing a collaborative-filtering based advertisement system for serving one or more advertisements to a current user of a network, the method comprising:

determining a rating of at least one entertainment item from the current user; and

selecting the one or more advertisements to serve to the current user using the at least one rating of the current user and a model based on ratings of a plurality of entertainment items and a plurality of advertisements from a plurality of previous network users.

2. The method of claim 1, wherein:

the model comprises a plurality of vectors, each vector comprising a rating of at least one entertainment item and a rating of at least one advertisement from a previous network user; and

selecting the one or more advertisements comprises;

determining a representative vector in the model having a corresponding rating of the at least one entertainment item that is most similar to the rating of the at least one entertainment item from the current user; and

examining a rating of the at least one advertisement in the representative vector to select the one or more advertisements.

3. The method of claim 1, wherein:

the model comprises a plurality of cluster vectors, each cluster vector comprising a center rating of at least one entertainment item, a variation from the center rating of the at least one entertainment item, and a center rating of at least one advertisement, each cluster vector representing ratings from two or more previous network users; and

selecting the one or more advertisements comprises;

determining a representative cluster vector in the model comprising a corresponding center rating of the at least one entertainment item and a variation from the center rating of the at least one entertainment item that contains the rating of the at least one entertainment item from the current user; and

examining a center rating of the at least one advertisement in the representative cluster vector to select the one or more advertisements.

4. The method of claim 1, wherein an entertainment item comprises items related to music, music videos, television programs, radio programs, movies, or video games.

5. The method of claim 1, wherein the one or more selected advertisements are not related in topic or subject to an entertainment item.

6. The method of claim 1, wherein:

a rating comprises an explicit rating or an implicit rating derived from a user event.

7. The method of claim 1, wherein:

the ratings of the plurality of entertainment items and the plurality of advertisements from the plurality of previous network users are normalized to a same scale.

8. The method of claim 1, further comprising determining at least one piece of demographic information of the current user, wherein:

the model is further based on demographic information of the plurality of previous network users; and

selecting the one or more advertisements comprises using the at least one piece of demographic information of the current user, the rating of the at least one entertainment item from the current user, and the model to select the one or more advertisements to serve to the current user.

9. A computer readable medium having instructions stored thereon when executed, implement a collaborative-filtering based advertisement system for serving one or more advertisements to a current user of a network, the computer readable medium comprising sets of instructions for:

determining a rating of at least one entertainment item from the current user; and

selecting the one or more advertisements to serve to the current user using the at least one rating of the current user and a model based on ratings of a plurality of entertainment items and a plurality of advertisements from a plurality of previous network users.

10. The computer readable medium of claim 9, wherein:

the model comprises a plurality of vectors, each vector comprising a rating of at least one entertainment item and a rating of at least one advertisement from a previous network user; and

the set of instructions for selecting the one or more advertisements comprises sets of instructions for;

determining a representative vector in the model having a corresponding rating of the at least one entertainment item that is most similar to the rating of the at least one entertainment item from the current user; and

examining a rating of the at least one advertisement in the representative vector to select the one or more advertisements.

11. The computer readable medium of claim 9, wherein:

the model comprises a plurality of cluster vectors, each cluster vector comprising a center rating of at least one entertainment item, a variation from the center rating of the at least one entertainment item, and a center rating of at least one advertisement, each cluster vector representing ratings from two or more previous network users; and

the set of instructions for selecting the one or more advertisements comprises sets of instructions for;

determining a representative cluster vector in the model comprising a corresponding center rating of the at least one entertainment item and a variation from the center rating of the at least one entertainment item that contains the rating of the at least one entertainment item from the current user; and

examining a center rating of the at least one advertisement in the representative cluster vector to select the one or more advertisements.

12. The computer readable medium of claim 9, wherein an entertainment item comprises items related to music, music videos, television programs, radio programs, movies, or video games.

13. The computer readable medium of claim 9, wherein the one or more selected advertisements are not related in topic or subject to an entertainment item.

14. The computer readable medium of claim 9, wherein:

a rating comprises an explicit rating or an implicit rating derived from a user event.

15. The computer readable medium of claim 9, wherein:

the ratings of the plurality of entertainment items and the plurality of advertisements from the plurality of previous network users are normalized to a same scale.

16. The computer readable medium of claim 9, further comprising a set of instructions for determining at least one piece of demographic information of the current user, wherein:

the model is further based on demographic information of the plurality of previous network users; and

the set of instructions for selecting the one or more advertisements comprises a set of instructions for using the at least one piece of demographic information of the current user, the rating of the at least one entertainment item from the current user, and the model to select the one or more advertisements to serve to the current user.

17. A system for implementing a collaborative-filtering based advertisement system for serving one or more advertisements to a current user of a network, the system comprising

a module configured for:

determining a rating of at least one entertainment item from the current user; and

selecting the one or more advertisements to serve to the current user using the at least one rating of the current user and a model based on ratings of a plurality of entertainment items and a plurality of advertisements from a plurality of previous network users.

18. The system of claim 17, wherein:

the model comprises a plurality of vectors, each vector comprising a rating of at least one entertainment item and a rating of at least one advertisement from a previous network user; and

the module is configured to select the one or more advertisements by;

determining a representative vector in the model having a corresponding rating of the at least one entertainment item that is most similar to the rating of the at least one entertainment item from the current user; and

examining a rating of the at least one advertisement in the representative vector to select the one or more advertisements.

19. The system of claim 17, wherein:

the model comprises a plurality of cluster vectors, each cluster vector comprising a center rating of at least one entertainment item, a variation from the center rating of the at least one entertainment item, and a center rating of at least one advertisement, each cluster vector representing ratings from two or more previous network users; and

the module is configured to select the one or more advertisements by;

determining a representative cluster vector in the model comprising a corresponding center rating of the at least one entertainment item and a variation from the center rating of the at least one entertainment item that contains the rating of the at least one entertainment item from the current user; and

examining a center rating of the at least one advertisement in the representative cluster vector to select the one or more advertisements.

20. The system of claim 17, wherein an entertainment item comprises items related to music, music videos, television programs, radio programs, movies, or video games.

21. The system of claim 17, wherein the one or more selected advertisements are not related in topic or subject to an entertainment item.

22. The system of claim 17, wherein:

a rating comprises an explicit rating or an implicit rating derived from a user event.

23. The system of claim 17, wherein:

the ratings of the plurality of entertainment items and the plurality of advertisements from the plurality of previous network users are normalized to a same scale.

24. The system of claim 17, wherein:

the module is further configured for determining at least one piece of demographic information of the current user;

the model is further based on demographic information of the plurality of previous network users; and

the module is configured to select the one or more advertisements by using the at least one piece of demographic information of the current user, the rating of the at least one entertainment item from the current user, and the model to select the one or more advertisements to serve to the current user.