Method and apparatus for generating recommendation scores using implicit and explicit viewing preferences

• US 8,843,965 B1
• Filed: 09/20/2000
• Issued: 09/23/2014
• Est. Priority Date: 09/20/2000
• Status: Expired due to Fees

• Current Assignee: Sisvel S.p.A. (Fineur International S.A.)
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Kurapati, Kaushal, Schaffer, J. David, Gutta, Srinivas
• Primary Examiner(s): DUFFIELD, JEREMY S
• Application Number:
  US09/666,401
• Time in Patent Office: 5,116 Days
• Field of Search: 725/34, 725/35, 725/46
• US Class Current: 725/46
• CPC Class Codes: H04N 21/25 Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies maintenance or administration in data networks H04L41/00
• CPC Class Codes: H04N 21/251 Learning process for intelligent management, e.g. learning user preferences for recommending movies details of learning user preferences for the retrieval of video data in a video database G06F16/739; computer systems using learning methods G06N3/08
• CPC Class Codes: H04N 21/45 Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
• CPC Class Codes: H04N 21/4532 involving end-user characteristics, e.g. viewer profile, preferences monitoring of user activities for profile generation for accessing a video database G06F16/739; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18
• CPC Class Codes: H04N 21/466 Learning process for intelligent management, e.g. learning user preferences for recommending movies services using the results of monitoring in broadcast systems H04H60/61
• CPC Class Codes: H04N 21/4667 Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
• CPC Class Codes: H04N 21/4668 for recommending content, e.g. movies
• CPC Class Codes: H04N 21/47 End-user applications
• CPC Class Codes: H04N 21/4755 for defining user preferences, e.g. favourite actors or genre retrieval personalisation and generation of user profiles for the retrieval of video data G06F16/739; user profiles in network data switching protocols H04L67/306
• CPC Class Codes: H04N 21/4756 for rating content, e.g. scoring a recommended movie
• CPC Class Codes: H04N 21/4826 using recommendation lists, e.g. of programs or channels sorted out according to their score
• CPC Class Codes: H04N 21/84 Generation or processing of descriptive data, e.g. content descriptors systems specially adapted for using meta-information in broadcast systems H04H60/73
• CPC Class Codes: H04N 7/16 Analogue secrecy systems; Analogue subscription systems

Abstract

A television programming recommender combines the explicit viewing preferences, i.e. user specified preferences, of viewers with their television viewing behavior (implicit preferences) to generate program recommendations. The explicit viewing preferences are utilized to generate an explicit recommendation score, E, for an upcoming television program. The implicit viewing preferences are utilized to generate an implicit recommendation score, I, for the program. A combined recommendation score, C, is computed based on the explicit and implicit scores, E and I. The combined recommendation score, C, can be computed using a weighted linear mapping.

Claims

1. A method for recommending television programs via a program recommending machine, comprising:

obtaining, by the machine, a list of one or more programs;

determining, by the machine, an explicit recommendation score, E, for the one or more programs based on an explicit profile associated with a user;

determining, by the machine, an implicit recommendation score, I, for the one or more programs based on an implicit profile associated with the user;

generating, by the machine, a combined recommendation score, C, based on the explicit recommendation score, E, and the implicit recommendation score, I; and

providing, by the machine, a recommendation of one or more of the programs in the list based on the combined recommendation score, C,

wherein said combined recommendation score, C, is substantially equally biased between said explicit recommendation score, E and the implicit recommendation score, I, when the explicit recommendation score, E, is substantially equal to the implicit recommendation score, I, and is biased increasingly towards said explicit recommendation score, E, in response to an increase in an absolute difference between the explicit recommendation score, E, and the implicit recommendation score, I.

2. The method of claim 1, wherein said list of one or more programs is obtained from an electronic program guide.

3. The method of claim 1, wherein said explicit recommendation score, E, is provided by an explicit program recommender.

4. The method of claim 1, wherein said explicit recommendation score, E, is defined as a weighted average of individual ratings of program features.

5. The method of claim 1, wherein said implicit recommendation score, I, is provided by an implicit program recommender.

6. The method of claim 1, wherein said combined recommendation score, C, is computed using a weighted linear mapping of said explicit recommendation score, E, and said implicit recommendation score, I.

7. The method of claim 1, including presenting said combined recommendation score, C, for each of said one or more programs to a user.

22. The method of claim 1, wherein the explicit recommendation score represents a complete explicit recommendation by an explicit program recommender, based on a first set of ratings of program features derived from an explicit profile;

the implicit recommendation score represents a complete implicit recommendation by an implicit program recommender, based on a second set of ratings of program features derived from an implicit profile; and

the generating of the combined recommendation score uses the implicit and explicit recommendation scores without reference to either the first or second set of ratings of program features.

26. The method of claim 1, wherein the explicit recommendation score, E, is normalized and is defined as a weighted average of individual ratings of program features determined using a first group of weights, and further wherein the combined recommendation score, C, is calculated in accordance with the following equation:

C={W_—

e*E+W_—

i*I)}/{W_—

e+W_—

i}

where W_e, and W_i are weighting coefficients of another group of weights and are operable to specify how much E and I contribute to the combined recommendation score, C, respectively, where W_e and W_i are substantially equal to each other when the explicit recommendation score E, is substantially equal to the implicit recommendation score I, and wherein W_e increases relative to W_i based upon an absolute difference between the explicit recommendation score E, and the implicit recommendation score I.

30. The method as claimed in claim 26, wherein the weights W_e and W_i are varied in dependence on a difference between E and I, where W_e≈

W_i, when E≈

I, and W_e>

>

W_i when either E>

I or E<

<

I.

8. A method for generating a recommendation score for a television program via a program recommending machine, comprising:

determining, at the machine, an explicit recommendation score, E, for the program based on an explicit profile associated with a user;

determining, at the machine, an implicit recommendation score, I, for the program based on an implicit profile associated with the user;

generating, at the machine, a combined recommendation score, C, for the program based on the explicit recommendation score, E, and the implicit recommendation score, I; and

selectively providing, at the machine, a recommendation of the program based on the combined recommendation score, C,

wherein said combined recommendation score, C, is substantially equally biased between said explicit recommendation score, E and the implicit recommendation score, I, when the explicit recommendation score, E, is substantially equal to the implicit recommendation score, I, and is biased increasingly towards said explicit recommendation score, E, in response to an increase in an absolute difference between the explicit recommendation score, E, and the implicit recommendation score, I.

9. The method of claim 8, wherein said program is identified in an electronic program guide.

10. The method of claim 8, wherein said explicit recommendation score, E, is provided by an explicit program recommender.

11. The method of claim 8, wherein said explicit recommendation score, E, is defined as a weighted average of individual ratings of program features.

12. The method of claim 8, wherein said implicit recommendation score, I, is provided by an implicit program recommender.

13. The method of claim 8, wherein said combined recommendation score, C, is computed using a weighted linear mapping of said explicit recommendation score, E, and said implicit recommendation score, I.

14. The method of claim 8, including presenting said combined recommendation score, C, for said program to a user.

23. The method of claim 8, wherein the explicit recommendation score represents a complete explicit recommendation by an explicit program recommender, based on a first set of ratings of program features derived from an explicit profile;

the implicit recommendation score represents a complete implicit recommendation by an implicit program recommender, based on a second set of ratings of program features derived from an implicit profile; and

the generating of the combined recommendation score uses the implicit and explicit recommendation scores without reference to either the first or the second set of ratings of program features.

27. The method of claim 8, wherein the combined recommendation score is calculated in accordance with the following equation:

C={W_—

e*E+W_—

i*I}/{W_—

e+W_—

i}

where W_e, and W_i are weighting coefficients operable to specify how much E and I contribute to the combined recommendation score, C, respectively.

31. The method as claimed in claim 27, wherein the weights W_e and W_i are varied in dependence on a difference between E and I, where W_e≈

W_i, when E≈

I, and W_e>

>

W_i when either E>

>

I or E<

<

I.

15. A system for obtaining a recommendation for a television program for a user, the program having attributes, comprising:

a memory for storing computer readable code; and

a processor operatively coupled to the memory, the processor configured to;

obtain a list of one or more programs;

determine an explicit recommendation score, E, for the one or more programs based on an explicit profile of a user;

determine an implicit recommendation score, I, for the one or more programs based on an implicit profile of the user;

generate a combined recommendation score, C, based on the explicit recommendation score, E, and the implicit recommendation score, I; and

provide a recommendation of one or more of the programs in the list based on the combined recommendation score,

wherein said combined recommendation score, C, is substantially equally biased between said explicit recommendation score, E and the implicit recommendation score, I, when the explicit recommendation score, E, is substantially equal to the implicit recommendation score, I, and is biased increasingly towards said explicit recommendation score, E, in response to an increase in an absolute difference between the explicit recommendation score, E, and the implicit recommendation score, I.

16. The system of claim 15, wherein said list of one or more programs is obtained from an electronic program guide.

17. The system of claim 15, wherein said explicit recommendation score, E, is defined as a weighted average of individual ratings of program features.

18. The system of claim 15, wherein said combined recommendation score, C, is computed using a weighted linear mapping of said explicit recommendation score, E, and said implicit recommendation score, I.

19. The system of claim 15, wherein said processor is further configured to present said combined recommendation score, C, for each of said one or more programs to a user.

24. The system of claim 15, wherein the explicit recommendation score represents a complete explicit recommendation by an explicit program recommender, based on a first set of ratings of program features derived from an explicit profile;

the implicit recommendation score represents a complete implicit recommendation by an implicit program recommender, based on a second set of ratings of program features derived from an implicit profile; and

the generating of the combined recommendation score uses the implicit and explicit recommendation scores without reference to either the first or the second set of ratings of program features.

28. The system of claim 15, wherein the combined recommendation score is calculated in accordance with the following equation:

C={W_—

e*E+W_—

i*I}/{W_—

e+W_—

i}

where W_e, and W_i are weighting coefficients operable to specify how much E and I contribute to the combined recommendation score, C, respectively.

32. The method as claimed in claim 28, wherein the weights W_e and W_i are varied in dependence on a difference between E and I, where W_e≈

W_i, when E≈

I and W_e>

>

W_i when either E>

>

I or E<

<

I.

20. A non-transitory computer-readable medium, readable by a data processing device, encoded with a computer program having code for causing the data processing device to perform the following method steps:

maintaining an explicit viewer profile with respect to a user;

maintaining an implicit viewer profile, with respect to the user, wherein the implicit user profile is distinct from the explicit viewer profile;

receiving information regarding at least one program;

generating an explicit recommendation score for the program based on the explicit viewer profile and at least part of the information regarding the program;

generating an implicit recommendation score for the program based on the implicit viewer profile and at least part of the information regarding the program;

calculating a combined recommendation score from the explicit recommendation score and the implicit recommendation score; and

providing a recommendation of the at least one program based on the combined recommendation score,

wherein said combined recommendation score, C, is substantially equally biased between said explicit recommendation score, E and the implicit recommendation score, I, when the explicit recommendation score, E, is substantially equal to the implicit recommendation score, I, and is biased increasingly towards said explicit recommendation score, E, in response to an increase in an absolute difference between the explicit recommendation score, E, and the implicit recommendation score, I.

21. The non-transitory computer-readable medium of claim 20, wherein said calculating includes taking a weighted average of the explicit recommendation score and the implicit recommendation score.

25. The non-transitory computer-readable medium of claim 20, wherein the explicit recommendation score represents a complete explicit recommendation by an explicit program recommender, based on a first set of ratings of program features derived from the explicit profile;

the implicit recommendation score represents a complete implicit recommendation by an implicit program recommender, based on a second set of ratings of program features derived from the implicit profile; and

the generating of the combined recommendation score uses the implicit and explicit recommendation scores without reference to either the first or second set of ratings of program features.

29. The medium of claim 20, wherein the combined recommendation score is calculated in accordance with the following equation:

C={W_—

e*E+W_—

i*I}/{W_—

e+W_—

i}

where W_e, and W_i are weighting coefficients operable to specify how much E and I contribute to the combined recommendation score, C, respectively.

33. The method as claimed in claim 29, wherein the weights W_e and W_i are varied in dependence on a difference between E and I, where W_e≈

W_i, when E≈

I, and W_e>

>

W_i when either E>

>

I or E<

<

I.