Subscriber identification system

• US 6,684,194 B1
• Filed: 12/02/1999
• Issued: 01/27/2004
• Est. Priority Date: 12/03/1998
• Status: Expired due to Term

• Current Assignee: Prime Research Alliance E, LLC
• Original Assignee: Expanse Networks, Inc.
• Inventors: Eldering, Charles A., Sylla, M. Lamine
• Primary Examiner(s): Stamber, Eric W.
• Assistant Examiner(s): Alvarez, Raquel
• Application Number:
  US09/452,893
• Time in Patent Office: 1,517 Days
• Field of Search: 705/1, 705/7, 705/10, 705/14, 345/745, 386/83, 725/9, 725/10, 725/42, 725/35
• US Class Current: 705/7.33
• 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 30/0204 Market segmentation
• CPC Class Codes: G06Q 30/0254 based on statistics
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 30/0277 Online advertisement
• CPC Class Codes: H04N 21/252 Processing of multiple end-users' preferences to derive collaborative data
• 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 21/4662 characterized by learning algorithms
• CPC Class Codes: H04N 21/6582 Data stored in the client, e.g. viewing habits, hardware capabilities, credit card number arrangements where receivers interact with the broadcast H04H20/38
• CPC Class Codes: H04N 7/17318 Direct or substantially direct transmission and handling of requests

Abstract

A subscriber identification system is presented in which subscriber selection data including channel changes, volume changes, and time-of-day viewing information is used to identify a subscriber from a group of subscribers. In one instance, the subscriber viewing data is recorded and a signal processing algorithm such as a Fourier transform is used to produce a processed version of the subscriber selection data. The processed version of the subscriber selection data can be correlated with stored common identifiers of subscriber profiles to determine which subscriber from the group is presently viewing the programming. A neural network or fuzzy logic can be used as the mechanism for identifying the subscriber from clusters of information which are associated with individual subscribers.

Claims

1. In a telecommunications environment comprising a delivery network for delivering content and a residential unit for receiving the content and interacting with a subscriber, the residential unit including at least a storage medium and a processor, a computer implemented method of associating a subscriber interacting with the residential unit with characteristics that are not directly related to interactions of the subscriber with the residential unit, the method comprising utilizing the processor to:

monitor subscriber interactions with the residential unit;

process at least a portion of the subscriber interactions to generate a session profile, wherein the session profile is generated without knowledge of the identity of the subscriber and identifies subscriber traits including traits that are not directly associated with the subscriber interactions;

compare at least a portion of the session profile with at least a portion of subscriber profiles stored in the storage medium, the subscriber profiles being a compilation of session profiles generated over multiple interaction sessions without knowledge of the identity of the subscriber that have been grouped together based on similarities, the subscriber profiles identifying subscriber characteristics that are a compilation of subscriber traits from the session profiles grouped together, the subscriber characteristics including characteristics that are not directly associated with the subscriber interactions; and

associate, based on similarities between the subscriber profile and the session profile, tile subscriber with one of the subscriber profiles and the subscriber characteristics including the characteristics that are not directly associated with the subscriber interactions.

2. The method of claim 1, wherein the subscriber traits and the subscriber characteristics include probabilistic demographics associated with the subscriber.

3. The method of claim 2, wherein said comparing includes comparing at least some subset of the probabilistic demographic characteristics of the subscriber within the session profile with similar attributes in the subscriber profiles.

4. The method of claim 1, wherein the subscriber traits and the subscriber characteristics include probabilistic program characteristics associated with the subscriber.

5. The method of claim 4, wherein said comparing includes comparing at least some subset of the probabilistic program characteristics associated with the subscriber within the session profile with similar attributes in the subscriber profiles.

6. The method of claim 1, wherein the subscriber traits and the subscriber characteristics include subscriber selection data.

7. The method of claim 6, wherein the subscriber selection data includes at least some subset of channel changes and volume changes.

8. The method of claim 6, wherein said comparing includes comparing at least some subset of the subscriber selection data within the session profile with similar attributes in the subscriber profiles.

9. The method of claim 1, wherein the subscriber traits and the subscriber characteristics include a subscriber signature.

10. The method of claim 9, wherein the subscriber signature is a discrete spectrum representation of at least some subset of the subscriber interactivity.

11. The method of claim 10, wherein the at least some subset of the subscriber interactivity used to generate the subscriber signature includes channel changes.

12. The method of claim 10, wherein the at least some subset of the subscriber interactivity used to generate the subscriber signature includes volume control.

13. The method of claim 9, wherein the subscriber signature is a Fourier transform of at least some subset of the subscriber interactivity.

14. The method of claim 9, wherein said comparing includes comparing the subscriber signature within the session profile with similar attributes in the subscriber profiles.

15. The method of claim 1, wherein said comparing is performed by a neural network.

16. The method of claim 1, wherein each subscriber profile represents a different device interactivity identity.

17. The method of claim 1, wherein the session profiles are clustered based on similar subscriber signatures in order to generate the subscriber profiles.

18. The method of claim 17, wherein the subscriber signatures are a discrete spectrum representation of at least some subset of the subscriber interactivity.

19. The method of claim 1, wherein the session profiles are clustered based on similarities between at least some subset of probabilistic demographic characteristics, probabilistic program characteristics, and subscriber selection data in order to generate the subscriber profiles.

20. The method of claim 1, wherein the device is a television.

21. The method of claim 1, further comprising retrieving program data associated with content selected within the subscriber interactivity.

22. The method of claim 21, wherein the program data defines at least some subset of program characteristics and program demographics.

23. The method of claim 21, wherein the program data is obtained by context mining textual information associated with the selected content.

24. The method of claim 23, wherein the textual information includes at least some subset of an electronic program guide, closed caption data, and an HTML file.

25. The method of claim 1, further comprising clustering the session profiles based on similarities between at least one identified subscriber trait to form subscriber profiles.

26. The method of claim 1, wherein each of the subscriber profiles may be associated with one or more market segments, each market segment being associated with one or more of the plurality of subscriber traits.

27. The method of claim 26, further comprising associating the subscriber with at least one market segment.

28. A system for associating a subscriber interacting with a device with characteristics that are not directly associated with device interactions or the subscriber, the system comprising:

the device;

a subscriber interface device for interacting with the device;

a storage medium; and

a processor for;

monitoring subscriber interactivity with the device;

processing at least a portion of the subscriber interactivity to generate a session profile, wherein the session profile is generated with no knowledge of the identity of the subscriber and identifies subscriber traits including traits that are not directly associated with the subscriber interactivity;

comparing at least a portion of the session profile with at least a portion of subscriber profiles, the subscriber profiles being a compilation of session profiles generated over multiple interaction sessions with no knowledge of the identity of the subscriber that have been grouped together based on similarities, the subscriber profiles identifying subscriber characteristics that are a compilation of subscriber traits from the session profiles grouped together, the subscriber characteristics including characteristics that are not directly associated with the subscriber interactivity; and

associating, based on similarities between the subscriber profile and the session profile, the subscriber with one of the subscriber profiles and the subscriber characteristics including the characteristics that are not directly associated with the subscriber interactivity.

29. The system of claim 28, further comprising a delivery network for delivering content to the device responsive to said subscriber interface.

30. The system of claim 28, wherein said processor compares the session profile with the subscriber profiles by using a neural network.

31. The system of claim 28, wherein said processor generates a session profile that includes probabilistic demographic characteristics and compares at least some subset of the probabilistic demographic characteristics of the subscriber included within the session profile with similar attributes in the subscriber profiles.

32. A The system of claim 28, wherein said processor generates a session profile that includes probabilistic program characteristics and compares at least some subset of the probabilistic program characteristics of the subscriber included within the session profile with similar attributes in the subscriber profiles.

33. The system of claim 28, wherein said processor generates a session profile that includes subscriber selection data and compares at least some subset of the subscriber selection data included within the session profile with similar attributes in the subscriber profiles.

34. The system of claim 28, wherein said processor generates a session profile that includes a subscriber signature and compares the subscriber signature included within the session profile with similar attributes in the subscriber profiles.

35. The system of claim 34, wherein tile subscriber signature is a discrete spectrum representation of at least some subset of the subscriber interactivity.

36. The system of claim 28, wherein each of the subscriber profiles represents a different device interactivity identity.

37. The system of claim 28, wherein said processor clusters the session profiles based on similar subscriber signatures in order to generate the subscriber profiles.

38. The system of claim 28, wherein said processor clusters the session profiles based on similarities between at least some subset of probabilistic demographic characteristics, probabilistic program characteristics, and subscriber selection data in order to generate the subscriber profiles.

39. The system of claim 28, wherein said processor

retrieves program data associated with the content delivered from the content provider; and

generates the session profile based on the program data and the subscriber interactivity.

40. An apparatus for associating a subscriber interacting with a device with characteristics that are not directly associated with device interactions of the subscriber, the apparatus comprising:

a subscriber interface for receiving subscriber interactions;

a content interface for receiving content from a content provider responsive to said subscriber interface;

a storage medium; and

a processor for;

processing at least a portion of the subscriber interactivity to generate a session profile, wherein the session profile is generated with no knowledge of the identity of the subscriber and identifies subscriber traits and at least some of the subscriber traits are discretionary traits that are not directly associated with the subscriber interactivity;

comparing at least a portion of the session profile with at least a portion of subscriber profiles, the subscriber profiles are generated with no knowledge of the identity of the subscriber and are a compilation of session profiles generated over multiple interaction sessions that have been grouped together based on similarities, the subscriber profiles identifying subscriber characteristics that are a compilation of subscriber traits from the session profiles grouped together, the subscriber characteristics including characteristics that are not directly associated with the subscriber interactivity; and

associating, based on similarities between the subscriber profiles and the session profile, the subscriber with one of the subscriber profiles and the subscriber characteristics including the characteristics that are not directly associated with the subscriber interactivity.

41. The apparatus of claim 40, wherein said processor compares the session profile with subscriber profiles by using a neural network.

42. The apparatus of claim 40, wherein said processor

generates a session profile that includes at least some subset of probabilistic demographic characteristics, probabilistic program characteristics, subscriber selection data and subscriber signature; and

compares attributes within the session profile with similar attributes in the subscriber profiles.

43. The apparatus of claim 40, wherein said processor

retrieves program data associated with the content received from the content provider, and

generates the session profile based on the program data and the subscriber interactions.

44. An apparatus for associating a subscriber interacting with a device with characteristics that are not directly associated with device interactions of the subscriber, the apparatus comprising:

a subscriber interface for receiving subscriber interactions;

a content interface for delivering content to subscribers;

a storage medium; and

a processor for;

processing at least a portion of the subscriber interactivity to generate a session profile, wherein the session profile is generated with no knowledge of the identity of the subscriber and identifies subscriber traits and at least some of the subscriber traits are discretionary traits that are not directly associated with the subscriber interactivity;

comparing at least a portion of the session profile with at least a portion of subscriber profiles, the subscriber profiles being a compilation of session profiles generated over multiple interaction sessions that have been grouped together based on similarities, the subscriber profiles identifying subscriber characteristics that are a compilation of subscriber traits from the session profiles grouped together, the subscriber characteristics including characteristics that are not directly associated with the subscriber interactivity; and

associating, based on similarities between the subscriber profiles and the session profile, the subscriber with one of the subscriber profiles and the subscriber characteristics including the characteristics that are not directly associated with the subscriber interactivity.

45. The apparatus of claim 44, wherein said processor

generates a session profile that includes at least some subset of probabilistic demographic characteristics, probabilistic program characteristics, subscriber selection data and subscriber signature; and

compares attributes within die session profile with similar attributes in the subscriber profiles.

46. The apparatus of claim 44, wherein said processor

retrieves program data associated with content selected by the subscriber;

and

generates the session profile based on the program data and the subscriber interactions.

47. A computer program embodied on a computer-readable medium for associating a subscriber interacting with a device with characteristics that are not directly associated with device interactions of the subscriber, said computer program comprising:

a source code segment for monitoring subscriber interactivity;

a source code segment for processing at least a portion of the subscriber interactivity to generate a session profile, wherein the session profile is generated with no knowledge of the identity of the subscriber and identifies subscriber traits and at least some of the subscriber traits are discretionary traits that are not directly associated with the subscriber interactivity;

a source code segment for comparing at least a portion of the session profile with at least a portion of subscriber profiles, the subscriber profiles being a compilation of session profiles generated over multiple interaction sessions that have been grouped together based on similarities, the subscriber profiles identifying subscriber characteristics that are a compilation of subscriber traits from the session profiles grouped together, the subscriber characteristics including characteristics that are not directly associated with the subscriber interactivity; and

a source code segment for associating, responsive to said source code segment for comparing, the subscriber with one of the subscriber profiles and the subscriber characteristics including the characteristics that are not directly associated with the subscriber interactivity.

48. The computer program of claim 47, wherein said source code segment for processing the subscriber interactions includes:

a source code segment for extracting program data from content selected as part of the subscriber interactivity; and

a source code segment for generating the session profile based on the subscriber interactivity and the program data.

49. The computer program of claim 47, wherein said source code segment for processing includes a source code segment for processing volume and channel change commands monitored as part of the subscriber interactions to generate a subscriber signature that is part of the session profile.

50. The computer program of claim 49, wherein said source code segment for processing volume and channel change commands utilizes a Fourier transform.

51. The computer program of claim 47, wherein the device is a television.

52. The computer program of claim 51, wherein the computer program is implemented within a set-top box connected to the television.

53. The computer program of claim 51, wherein the computer program is implemented within a television access network.

54. The computer program of claim 47, wherein the computer program is implemented as a client-server architecture.

55. The a telecommunications environment comprising a delivery network for delivering content and a residential unit for receiving the content and interacting with a subscriber, the residential unit including at least a storage medium and a processor, a computer implemented system for associating a subscriber interacting with the residential unit with characteristics that are not directly associated with interactions of the subscriber with the residential unit, the system comprising:

means for monitoring subscriber interactions;

means for processing at least a portion of the subscriber interactions to generate a session profile, wherein the session profile is generated without knowledge of the identity of the subscriber and identifies subscriber traits and at least some of the subscriber traits are discretionary traits that are not directly associated with the subscriber interactions, wherein said means for processing utilizes the processor;

means for comparing at least a portion of the session profile with at least a portion of subscriber profiles stored in the storage medium, the subscriber profiles being a compilation of session profiles generated over multiple interaction sessions that have been grouped together based on similarities, the subscriber profiles identifying subscriber characteristics that are a compilation of subscriber traits from the session profiles grouped together, the subscriber characteristics including characteristics that are not directly associated with the subscriber interactions; and

means for associating, responsive to said means for comparing, the subscriber with one of the subscriber profiles and the subscriber characteristics including the characteristics that are not directly associated with the subscriber interactions.

56. The system of claim 55, wherein said means for processing the subscriber interactions includes:

means for extracting program data from content selected as part of the subscriber interactivity, and

means for generating the session profile based on the subscriber interactivity and the program data.

57. The system of claim 55, wherein said means for processing includes means for processing volume and channel change commands monitored as part of the subscriber interactions to generate a subscriber signature that is part of the session profile.

58. The system of claim 57, wherein said means for processing volume and channel change commands utilizes a Fourier transform.

59. In a television network environment consisting of a display device, a storage medium, and a processor, a computer implemented method for associating traits to a television subscriber, the traits not being directly related to interactions of the television subscriber with a television, the method comprising:

monitoring subscriber interactions with the television for a plurality of viewing sessions;

utilizing the processor to process, for each viewing session, at least a portion of the subscriber interactions to create session profiles that identify traits about the subscriber including traits not directly associated with the interactions;

wherein the session profiles do not include any data related to actual subscriber identity;

clustering the session profiles based on similarities between at least some portion of the session profiles to form subscriber profiles, wherein the subscriber profiles do not include any data related to actual subscriber identity and identity a plurality of traits that were identified in the clustered session profiles, the plurality of traits including traits not directly associated with the interactions;

comparing, for each viewing session, at least some portion of the session profile with a corresponding portion of the subscriber profiles,

selecting, for each viewing session, a subscriber profile for the subscriber based on said comparing; and

associating the plurality of traits associated with the subscriber profile to the subscriber including the traits not directly associated with the interactions.

60. The method of claim 59, wherein the plurality of traits include demographic traits.

61. In a television network environment consisting of a display device, a storage medium, and a processor, a computer implemented method for associating traits to a television subscriber, the traits not being directly related to interactions of the television subscriber with a television, the method comprising:

monitoring subscriber interactions with the television for a plurality of viewing sessions;

utilizing the processor to process, for each viewing session, at least a portion of the subscriber interactions to create session profiles that identify traits about the subscriber including traits not directly associated with the interactivity, wherein the session profiles are generated with no knowledge of subscriber identity;

clustering the session profiles based on similarities between at least some portion of the session profiles to form subscriber profiles, wherein the subscriber profiles are generated without knowledge of subscriber identity and identify a plurality of traits that were identified in the clustered session profiles, the plurality of traits including traits not directly associated with the interactions;

comparing, for each viewing session, at least some portion or the session profile with a corresponding portion of the subscriber profiles,

selecting, for each viewing session, a subscriber profile for the subscriber based on said comparing; and

associating the plurality of traits associated with the subscriber profile to the subscriber including the traits not directly associated with the interactions.