Scoring and recommending media content based on user preferences

• US 20030225777A1
• Filed: 05/31/2002
• Published: 12/04/2003
• Est. Priority Date: 05/31/2002
• Status: Abandoned Application

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Marsh, David J.
• Application Number:
  US10/160,932
• Publication Number:
  US 20030225777A1
• Time in Patent Office: Days
• US Class Current: 707/101
• CPC Class Codes: H04N 21/44222 Analytics of user selections, e.g. selection of programs or purchase activity monitoring of user selections in data processing systems G06F11/34; arrangements for monitoring the user's behaviour or opinions in broadcast systems H04H60/33
• 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 21/4668 for recommending content, e.g. movies
• 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 21/84 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 7/163 by receiver means only

Abstract

Systems and methods are described for scoring and accurately recommending multimedia content programming to users based upon a user'"'"'s preferences, each user receiving individualized programming recommendations according to that user'"'"'s likes and dislikes. A user provides preferred values for attributes of television programs. For example, if the user likes reality shows, the user would assign a relatively high attribute score for a genre attribute having a value of ‘reality show.” The preferred values are compared to a program description file that list program attribute values for a program available for viewing. A program score is obtained based on this comparison. If there are many matches, then the program score will be high. Programs are recommended to the user based on the program scores of the programs; programs having higher program scores (from having many matches with the user'"'"'s preferences) will be recommended over lower-scoring programs.

Claims

1. A method, comprising:

comparing one or more user preference attribute values to one or more program attribute values associated with a multimedia program to identify one or more matches;

deriving an attribute score for each attribute for which a match is identified; and

calculating a program score from the attribute scores.

2. The method as recited in claim 1, wherein the calculating a program score further comprises calculating the program score by adding the attribute scores derived for the multimedia program.

3. The method as recited in claim 1, wherein:

the user preference attribute values are contained in a user preference file (UPF) that is uniquely associated with a user; and

the program attribute values are contained in a content description file (CDF) that is uniquely associated with the multimedia program.

4. The method as recited in claim 1, further comprising recommending a program based on program scores derived for the program.

5. The method as recited in claim 4, wherein the recommending further comprises recommending the program if a program score associated with the program is higher than a threshold score.

6. The method as recited in claim 4, wherein the recommending further comprises:

determining if there is sufficient storage space available in a content buffer that stores one or more recommended programs;

if there is sufficient memory available, storing the program regardless of the program score; and

if there is not sufficient memory available, deleting one or more stored programs having program scores lower than the program score for the program, and storing the program.

7. The method as recited in claim 1, wherein the multimedia program is a television program.

8. The method as recited in claim 1, wherein the comparing further comprises comparing each user preference attribute value with each program attribute value.

9. The method as recited in claim 1, wherein the calculating an attribute score further comprises:

identifying a preference rating associated with an attribute for which a match was found; and

assigning the preference rating as an attribute score for the attribute.

10. The method as recited in claim 1, wherein the calculating an attribute score further comprises:

identifying a preference rating associated with an attribute for which a match was identified;

identifying a significance value associated with the attribute; and

calculating the attribute score from the preference rating and the significance value.

11. The method as recited in claim 10, wherein the calculating the attribute score further comprises multiplying the preference rating by the significance value.

12. The method as recited in claim 10, wherein the preference rating ranges from −

5 to +5.

13. The method as recited in claim 10, wherein the significance value ranges from 0 to 100.

14. The method as recited in claim 1, wherein the user preference attribute values and the program attribute values conform to a content description schema.

15. A method for recommending content to a user, comprising:

comparing one or more user preference attribute values that are used to describe one or more of a user'"'"'s content preferences with a content description that includes one or more program attribute values;

assigning an attribute score to each user preference attribute found to have a value that matches a program attribute value;

deriving a content score from the attribute scores; and

determining whether or not to recommend the content to the user based on the attribute scores.

16. The method as recited in claim 15, further comprising storing the content in a content buffer if the content is recommended, wherein the content buffer is used to store recommended content.

17. The method as recited in claim 15, further comprising recommending content by providing a content description to the user and indicating that the content is recommended.

18. The method as recited in claim 15, wherein the deriving a content score further comprises assigning a user preference rating as an attribute score for an attribute found to have a value that matches a program attribute value, the user preference rating being assigned by the user to indicate how much the user values the attribute.

19. The method as recited in claim 15, wherein the deriving a content score further comprises:

identifying a user preference rating associated with a matched attribute, the user preference rating being assigned by the user to indicate how much the user values the matched attribute;

identifying a significance value associated with the matched attribute, the significance rating indicating a relative importance of the matched attribute to other attributes; and

scoring the matched attribute by applying a formula to the user preference rating and the significance value.

20. The method as recited in claim 19, wherein the scoring further comprises multiplying the user preference rating by the significance value to obtain an attribute score for the matched attribute.

21. The method as recited in claim 15, wherein the determining whether or not to recommend the content further comprises recommending the content if the content score is greater than or equal to a threshold score.

22. The method as recited in claim 15, wherein the determining whether or not to recommend the content further comprises:

storing the content in a content buffer if there is sufficient storage space available in the content buffer; and

if there is not sufficient memory available in the content buffer, deleting one or more stored content from the content buffer if the stored content has a lower content score than the content, and storing the content in the content buffer.

23. The method as recited in claim 15, wherein:

the user preference attribute values are defined in a user preference file (UPF) according to a content description schema; and

the content description is defined in a content description file (CDF) according to the content description schema.

24. The method as recited in claim 15, wherein the content further comprises a television program.

25. A method, comprising:

assigning a significance value to each of multiple program attributes associated with multiple content programs;

storing the significance values in a significance file; and

wherein the significance value associated with a program attribute denotes a relative importance of the program attribute as compared with other program attributes.

26. The method as recited in claim 25, wherein each program attribute has a unique significance value associated therewith.

27. The method as recited in claim 25, wherein the program attributes belong to a content description schema.

28. The method as recited in claim 25, wherein the significance values range from 0 to 100.

29. A method, comprising:

obtaining a set of user preference values that indicate program attribute values preferred by a user; and

storing the user preference values in a user preference file (UPF) that is uniquely associated with the user.

30. The method as recited in claim 29, further comprising providing a preference questionnaire to the user to obtain the set of user preferences.

31. The method as recited in claim 29, wherein the user preference values are stored in the UPF according to a content description schema.

32. The method as recited in claim 29, further comprising updating the UPF with new user preference values after the UPF has been created.

33. The method as recited in claim 32, wherein the updating occurs periodically at predefined intervals.

34. The method as recited in claim 32, wherein the updating occurs whenever the user provides the new user preference values.

35. The method as recited in claim 29, further comprising:

monitoring one or more program attributes of programs watched by the user; and

updating the UPF with at least one program attribute value of the program attributes, storing the program attribute value as a user preference.

36. The method as recited in claim 35, further comprising:

creating a user viewing log that stores program attribute values of the monitored program attributes; and

wherein the updating further comprises periodically updating the UPF with the program attribute values stored in the user viewing log.

37. A system, comprising:

a user preference file (UPF) uniquely associated with a user, the UPF storing one or more user preferences indicated by preference attribute values associated with program attributes of one or more programs;

a matching engine configured to;

compare the user preferences with program attribute values contained in a content description file (CDF), the program attribute values describing a program uniquely associated with the CDF;

identify program attributes having program attribute values in the CDF that match a preference attribute value in the UPF;

assign an attribute score to each program attribute having matching values in the CDF and the UPF; and

compute a program score for the program associated with the CDF, the program score being computed from the attribute scores.

38. The system as recited in claim 37, wherein the matching engine is further configured to recommend the program to the user if a recommendation condition is satisfied.

39. The system as recited in claim 38, wherein the recommendation condition further comprises the program score being equal to or greater than a threshold score.

40. The system as recited in claim 38, further comprising a content buffer for storing recommended programs, and wherein the recommendation condition further comprises the content buffer having sufficient storage space available to store the program.

41. The system as recited in claim 40, wherein the matching engine is further configured to delete one or more stored programs from the content buffer when there is insufficient memory available to store the program, if the program has a higher program score than the stored programs.

42. The system as recited in claim 37, further comprising a user preference file questionnaire that is presented to the user to provide user preference information for the UPF.

43. The system as recited in claim 37, further comprising a user viewing log generator configured to monitor programs viewed by the user and store program attribute values of the monitored programs in a user viewing log.

44. The system as recited in claim 43, further comprising a preference inference engine configured to determine new user preferences from the user viewing log and to update the UPF with the new user preferences.

45. The system as recited in claim 37, wherein the matching engine is further configured to compute the program score by summing the attribute scores derived for the program.

46. The system as recited in claim 37, wherein the matching engine is further configured to assign an attribute score by identifying a preference rating assigned to an attribute for which a matching value has been identified, and to assign the preference rating as the attribute score.

47. The system as recited in claim 37, further comprising a significance file that contains a significance value for each program attribute available for a program, and wherein the matching engine is further configured to:

assign an attribute score by identifying a preference rating assigned to an attribute for which a matching value has been identified, and to assign the preference rating as the attribute score; and

compute a program score from the attribute score of each attribute and the significance value associated with each respective attribute.

48. The system as recited in claim 47, wherein the matching engine is further configured to compute the program score by taking the sum of values derived from multiplying each attribute score by the associated significance value.

49. The system as recited in claim 37, wherein the user preferences in the UPF are defined according to a content description schema.

50. The system as recited in claim 49 wherein the program attributes in the CDF are defined according to the content description schema.

51. A system, comprising:

a preference file that stores preferred program attribute values that a user prefers in programs;

a matching engine configured to compare the preferred program attribute values with program description attribute values that are associated with a program available for viewing by the user, and to recommend the program to the user if the program meets a recommendation standard based on the comparisons.

52. The system as recited in claim 51, wherein the matching engine is further configured to:

calculate a program score based on the comparisons; and

recommend the program if the program score is at least as high as a scoring threshold.

53. The system as recited in claim 52, wherein the matching engine is further configured to calculate the program score by:

assigning an attribute score to each preferred program attribute value that matches a program description attribute value; and

summing the attribute scores to derive the program score.

54. The system as recited in claim 52, wherein the matching engine is further configured to calculate the program score by:

assigning an attribute score to each preferred program attribute value that matches a program description attribute value, the attribute score being an attribute preference rating pre-assigned for the preferred program attribute value; and

deriving a weighted attribute score by factoring a significance value into the attribute score, the significance value identifying a relative importance of the attribute as compared to other program description attribute value; and

summing the weighted attribute scores to derive the program score.

55. The system as recited in claim 51, wherein the preferred program attribute values and the program description attribute values are defined according to a content description schema which provides a standard for describing attributes of programs.

56. The system as recited in claim 51, further comprising a content buffer used to store recommended programs and wherein the matching engine is further configured to stored recommended programs in the content buffer.

57. The system as recited in claim 56, wherein the matching engine is further configured to store a program if there is sufficient buffer space available in the content buffer to store the program.

58. The system as recited in claim 56, wherein the matching engine is further configured to:

determine if there is sufficient buffer storage space available in the content buffer to store the program; and

if there is not sufficient space available to store the program, delete enough stored programs to make space available to store the program if the program is more highly recommended than the stored programs that are deleted, and to store the program in the content buffer.

59. One or more computer-readable media containing electronic representations of:

one or more preferred program attributes, each preferred program attribute identifying an attribute of a multimedia program; and

one or more attribute values associated with each of the preferred program attributes, the attribute values identifying a value preferred by a user to be available in a multimedia program.

60. The one or more computer-readable media as recited in claim 59, further comprising a preference rating associated with each attribute value, each preference rating indicating how much the user likes or dislikes the attribute value.

61. The one or more computer-readable media as recited in claim 60, wherein the preference ratings range from −

5 to +5.

62. The one or more computer-readable media as recited in claim 60, wherein the preference ratings range from −

3 to +3.

63. The one or more computer-readable media as recited in claim 59, wherein the one or more attribute values conform to a content description schema used to describe content programs.

64. One or more computer-readable media containing computer-executable instructions that, when executed on a computer, perform the following steps:

comparing preferred attribute values that identify program attributes preferred by a user to program attribute values associated with a program available for viewing by the user;

updating a program score associated with the program whenever a match is detected between a program attribute value and a preferred attribute value;

determining whether or not to recommend the program to the user based on the program score.

65. The one or more computer-readable media as recited in claim 64, further comprising instructions to store the program in a content buffer if the determination is made to recommend the program.

66. The one or more computer-readable media as recited in claim 65, further comprising instructions to provide a program description to the user and indicate that the program is recommended if a determination has been made to recommend the program.

67. The one or more computer-readable media as recited in claim 64, further comprising instructions to initialize the program score, a wherein the updating a program score further comprises adding a preference rating associated with the preferred attribute value that matches the program attribute value to the program score when the match is detected.

68. The one or more computer-readable media as recited in claim 67, further comprising instructions to multiply the preference rating for a preferred attribute by a significance value associated with an attribute to which the preferred attribute value corresponds before adding the preference rating to the program score.

69. The one or more computer-readable media as recited in claim 64, wherein the preferred attribute values and the program attribute values conform to a content description schema.

70. One or more computer-readable media containing computer-executable instructions that, when executed on a computer, perform the following steps:

comparing preferred attribute values associated with a user to program attribute values associated with a multimedia program available to a user;

recommending the multimedia program to the user if the comparisons satisfy one or more recommendation condition.

71. The one or more computer-readable media as recited in claim 70, wherein the preference attribute values are maintained in a user preference file (UPF).

72. The one or more computer-readable media as recited in claim 71, wherein the UPF is established with user responses to a UPF questionnaire.

73. The one or more computer-readable media as recited in claim 71, further comprising:

monitoring programs consumed by the user;

generating a user viewing log that outlines program attribute values associated with the consumed programs;

determining new preferred attribute values from the user viewing log; and

updating the UPF with the new preferred attribute values.

74. The one or more computer-readable media as recited in claim 70, wherein the program attribute values are maintained in a content description file that is uniquely associated with the multimedia program.

75. The one or more computer-readable media as recited in claim 70, wherein the preferred attribute values and the program attribute values conform to a content description schema.

76. The one or more computer-readable media as recited in claim 70, wherein the recommending further comprises:

providing a detailed program description of the multimedia program to the user; and

identifying the multimedia program as a recommended program.

77. The one or more computer-readable media as recited in claim 70, wherein the recommending further comprises storing the multimedia program in a content buffer that stores recommended programs if the multimedia program is recommended to the user, if there is sufficient memory available in the content buffer to store the multimedia program.

78. The one or more computer-readable media as recited in claim 77, further comprising:

computing a program score for the multimedia program based on the comparisons;

if the content buffer does not contain sufficient available memory to store the multimedia program in the content buffer, comparing the program score for the multimedia program with program scores for each of one or more stored programs contained in the content buffer;

deleting one or more stored programs that have program scores lower than the program score of the multimedia program to free sufficient storage space in the content buffer to store the multimedia program; and

storing the multimedia program in the content buffer.

79. The one or more computer-readable media as recited in claim 70, further comprising:

determining a preference rating for each attribute for which a preferred attribute value matches a program attribute value;

summing the preference ratings determined to derive a program score; and

wherein the recommendation condition is related to the program score.

80. The one or more computer-readable media as recited in claim 79, further comprising weighting each preference rating with a significance value associated with the attribute associated with the preference rating, the significance value identifying a relative importance of the attribute as compared to other attributes, said weighting performed prior to summing the preference ratings.

81. The one or more computer-readable media as recited in claim 70, wherein the multimedia program is a television program.

82. The one or more computer-readable media as recited in claim 70, wherein the multimedia program is an audio program.