Method and apparatus for identifying unique client users from user behavioral data

• US 7,092,926 B2
• Filed: 04/05/2002
• Issued: 08/15/2006
• Est. Priority Date: 04/06/2001
• Status: Active Grant

• Current Assignee: Cox Communications Incorporated (Cox Enterprises Incorporated)
• Original Assignee: Sedna Patent Services LLC
• Inventors: Cerrato, Dean E.
• Primary Examiner(s): Hirl, Joseph P.
• Application Number:
  US10/117,654
• Publication Number:
  US 20020178257A1
• Time in Patent Office: 1,593 Days
• Field of Search: 706/48, 706/14, 706/12, 434/350
• US Class Current: 706/48
• CPC Class Codes: G06F 21/316 by observing the pattern of computer usage, e.g. typical user behaviour
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: H04H 60/45 for identifying users

Abstract

A method and system are provided for identifying a current user of a terminal device from a group of possible users. The method includes providing a database containing multiple user input pattern profiles of prior user inputs to the terminal device. Each of the possible users of the group are associated with at least one of the user input pattern profiles. Current input patterns from use of the terminal device are detected. The current input patterns are combined and then dynamically matched with one of the user input pattern profiles, and the possible user associated with the matched user input pattern profile is selected as the current user. The system for identifying a current user of a terminal device from a group of possible users includes a database containing multiple user input pattern profiles of prior user inputs to the terminal device. Each of the possible users is associated with at least one of the user input pattern profiles. The system detects current input patterns from use of the terminal device, and then combines the patterns and dynamically matches the patterns with one of the user input pattern profiles. The system selects the possible user associated with the matched user input pattern profiles as the current user.

Claims

1. A method, comprising:

providing a database containing a plurality of user input pattern profiles of prior user inputs to a terminal device, each of a group of possible users of said terminal device being associated with at least one of said user input pattern profiles, said user input pattern comprising a clickstream;

detecting at least one current input pattern from use of said terminal device;

dynamically matching said at least one current input pattern with one of said user input pattern profiles; and

identifying a current user of said terminal device from said group of possible users by selecting the possible user associated with the one of said user input pattern profiles as the current user.

2. The method of claim 1 wherein said at least one current input pattern comprises a plurality of different input patterns, and wherein dynamically matching said at least one current input pattern comprises combining said plurality of different patterns and matching a combination of said different input patterns with one of said user input pattern profiles.

3. The method of claim 1 further comprising retraining said plurality of user input pattern profiles in said database with said at least one current input pattern.

4. The method of claim 1 further comprising determining a personal user profile associated with the current user.

5. The method of claim 4 further comprising transmitting targeted content to said current user in accordance with said personal user profile.

6. The method of claim 1 wherein said clickstream data relates to particular Web sites visited by the user or the duration of visits to the Web sites.

7. The method of claim 1 wherein said current input pattern comprises user keystroke data.

8. The method of claim 7 wherein said keystroke data comprises digraph interval data.

9. The method of claim 1 wherein said current input pattern comprises user mouse usage data.

10. The method of claim 1 wherein said current input pattern comprises user remote control usage data.

11. The method of claim 1 wherein said terminal device comprises a computer.

12. The method of claim 1 wherein said terminal device comprises a television set top box.

13. The method of claim 1 wherein said steps are implemented in a computer, and said computer communicates with said terminal device over a network.

14. The method of claim 13 wherein said network comprises the Internet.

15. The method of claim 13 wherein said network comprises a nodal television distribution network.

16. The method of claim 1 wherein said clickstream pattern relates to particular programs or channels selected by the user or the duration of viewing of said programs or channels.

17. A system, comprising:

a database containing a plurality of user input pattern profiles of prior user inputs to a terminal device, each of a group of possible users of said terminal device being associated with at least one of said user input pattern profiles, said user input pattern comprising a clickstream;

means for detecting at least one current input pattern from use of said terminal device;

means for dynamically matching said at least one current input pattern with one of said user input pattern profiles; and

identifying a current user of said terminal device from said group of possible users by selecting the possible user associated with the one of said user input pattern profiles as the current user.

18. The system of claim 17 wherein said at least one current input pattern comprises a plurality of different input patterns, and wherein said means for dynamically matching said at least one current input pattern combines said plurality of different patterns and matches a combination of said different input patterns with one of said user input pattern profiles.

19. The system of claim 17 further comprising means for retraining said plurality of user input pattern profiles in said database with said at least one current input pattern.

20. The system of claim 17 further comprising means for determining a personal user profile associated with the current user.

21. The system of claim 20 further comprising means for transmitting targeted content to said current user in accordance with said personal user profile.

22. The system of claim 17 wherein said clickstream data relates to particular Web sites visited by the user or the duration of visits to the Web sites.

23. The system of claim 17 wherein said current input pattern comprises user keystroke data.

24. The system of claim 23 wherein said keystroke data comprises digraph interval data.

25. The system of claim 17 wherein said current input pattern comprises user mouse usage data.

26. The system of claim 17 wherein said current input pattern comprises user remote control usage data.

27. The system of claim 17 wherein said terminal device comprises a computer.

28. The system of claim 17 wherein said terminal device comprises a television set top box.

29. The system of claim 17 wherein said system is implemented in a computer, and said computer communicates with said terminal device over a network.

30. The system of claim 29 wherein said network comprises the Internet.

31. The system of claim 29 wherein said network comprises a nodal television distribution network.

32. The system of claim 17 wherein said clickstream data relates to particular programs or channels selected by the user or the duration of viewing of said programs or channels.

33. A computer system, comprising:

memory for storing a program and a plurality of user input pattern profiles of prior user inputs to a terminal device, each of a group of possible users of said terminal device being associated with at least one of said user input pattern profiles, said user input pattern comprising a clickstream; and

a processor operative with the program to;

(a) detect at least one current input pattern from use of said terminal device;

(b) dynamically match said at least one current input pattern with one of said user input pattern profiles; and

(c) identifying a current user of said terminal device from said group of possible users by selecting the possible user associated with the one of said user input pattern profiles as the current user.

34. A method, comprising:

providing a database containing a plurality of user input pattern profiles of prior user inputs to a terminal device, each of a plurality of possible users of said terminal device being associated with at least one of said user input pattern profiles said user input pattern comprising a clickstream;

detecting at least one current input pattern from use of said terminal device;

dynamically matching said at least one current input pattern with one of said user input pattern profiles;

identifying a current user of said terminal device from said possible users by selecting the possible user associated with the one of said user input pattern profiles as the current user;

determining a personal user profile associated with the current user; and

transmitting targeted content to said current user in accordance with said personal user profile.

35. The method of claim 34 wherein said targeted content comprises targeted advertising.

36. The method of claim 34 wherein said targeted content comprises recommended program viewing choices.

37. The method of claim 34 wherein said personal profile includes demographic or preference data on said current user.

38. The method of claim 37 wherein said demographic or preference data includes data on at least one of user age, user sex, number of children, income, and geographic location.

39. The method of claim 34 wherein said steps are implemented in a computer server.

40. The method of claim 39 wherein said server comprises a video server.

41. The method of claim 39 wherein said server comprises a Web server.

42. The method of claim 34 wherein said terminal device comprises a set top box and a television monitor.

43. The method of claim 34 wherein said terminal device comprises a personal computer.

44. A method, comprising:

detecting a plurality of different types of current input patterns from use of a terminal device by a current user, said input pattern including a clickstream;

performing a soft match of each of said plurality of different types of current input patterns with a plurality of stored input patterns for each of said types of input patterns, said stored patterns representing input patterns for a group of possible users of said terminal device, said soft matches generating scored possible matches for each of said different types of data;

determining possible combinations of said scored possible matches;

determining a score for each said combination; and

identifying said current user of said terminal device from said group of possible users by selecting a possible user associated with said combination as the current user for the combination having the highest score.

45. The method of claim 44 further comprising retraining said plurality of stored input patterns with said current input patterns.

46. The method of claim 44 further comprising determining a personal user profile associated with the current user.

47. The method of claim 46 further comprising transmitting targeted content to said current user in accordance with said personal user profile.

48. The method of claim 44 wherein said clickstream pattern relates to particular Web sites visited by the user or the duration of visits to the Web sites.

49. The method of claim 44 wherein said different types of current input patterns include a user keystroke pattern.

50. The method of claim 49 wherein said keystroke pattern includes digraph interval data.

51. The method of claim 44 wherein said different types of current input patterns include user mouse usage data.

52. The method of claim 44 wherein said different types of current input patterns include user remote control usage data.

53. The method of claim 44 wherein said terminal device comprises a computer.

54. The method of claim 44 wherein said terminal device comprises a television set top box.

55. The method of claim 44 wherein said steps are implemented in a computer, and said computer communicates with said terminal device over a network.

56. The method of claim 55 wherein said network comprises the Internet.

57. The method of claim 55 wherein said network comprises a nodal television distribution network.

58. The method of claim 44 wherein said clickstream pattern relates to particular programs or channels selected by the user or the duration of viewing of said programs or channels.

59. A system, comprising:

means for detecting a plurality of different types of current input patterns from use of a terminal device by a current user, said input patterns including a clickstream;

means for performing a soft match of each of said plurality of different types of current input patterns with a plurality of stored input patterns for each of said types of input patterns, said stored patterns representing input patterns for a group of possible users of said terminal device, said soft matches generating scored possible matches for each of said different types of data;

means for determining possible combinations of said scored possible matches;

means for determining a score for each said combination; and

means for identifying said current user of said terminal device from said group of possible users by selecting a possible user associated with combination having the highest score as the current user.

60. The system of claim 59 further comprising means for retraining said plurality of stored input patterns with said current input patterns.

61. The system of claim 59 further comprising means for determining a personal user profile associated with the current user.

62. The system of claim 59 further comprising means for transmitting targeted content to said current user in accordance with said personal user profile.

63. The system of claim 59 wherein said clickstream pattern relates to particular Web sites visited by the user or the duration of visits to the Web sites.

64. The system of claim 59 wherein said different types of current input patterns include a user keystroke pattern.

65. The system of claim 64 wherein said keystroke pattern includes digraph interval data.

66. The system of claim 59 wherein said different types of current input patterns include user mouse usage data.

67. The system of claim 59 wherein said different types of current input patterns include user remote control usage data.

68. The system of claim 59 wherein said terminal device comprises a computer.

69. The system of claim 59 wherein said terminal device comprises a television set top box.

70. The system of claim 59 wherein said system is implemented in a computer, and said computer communicates with said terminal device over a network.

71. The system of claim 70 wherein said network comprises the Internet.

72. The system of claim 70 wherein said network comprises a nodal television distribution network.

73. The system of claim 59 wherein said clickstream pattern relates to particular programs or channels selected by the user or the duration of viewing of said programs or channels.

74. A method, comprising:

providing a database containing a plurality of user input pattern profiles of prior user inputs to a terminal device, various subsets of a set of possible users of said terminal device being associated with at least one of said user input pattern profiles, said user input pattern including a clickstream;

detecting at least one current input pattern from use of said terminal device by a current subset of users;

dynamically matching said at least one current input pattern with one of said user input pattern profiles; and

identifying said current subset of users of said terminal device from said set of possible users by selecting the subset of users associated with the one of said user input pattern profiles as the current subset of users.

75. The method of claim 74 wherein said at least one current input pattern comprises a plurality of different input patterns, and wherein dynamically matching said at least one current input pattern comprises combining said plurality of different patterns and matching a combination of said different input patterns with one of said user input pattern profiles.

76. The method of claim 74 further comprising retraining said plurality of user input pattern profiles in said database with said at least one current input pattern.

77. The method of claim 74 further comprising determining a personal user profile associated with the current subset of users.

78. The method of claim 77 further comprising transmitting targeted content to said terminal device in accordance with said personal user profile.

79. The method of claim 74 wherein said clickstream data relates to particular Web sites visited by the current subset of users or the duration of visits to the Web sites.

80. The method of claim 74 wherein said clickstream data relates to particular programs or channels selected by the subset of users or the duration of viewing of said programs or channels.

81. The method of claim 74 wherein said current input pattern comprises user keystroke data.

82. The method of claim 81 wherein said keystroke data comprises digraph interval data.

83. The method of claim 74 wherein said current input pattern comprises user mouse usage data.

84. The method of claim 74 wherein said current input pattern comprises user remote control usage data.

85. The method of claim 74 wherein said terminal device comprises a computer.

86. The method of claim 74 wherein said terminal device comprises a television set top box.

87. The method of claim 74 wherein said steps are implemented in a computer, and said computer communicates with said terminal device over a network.

88. The method of claim 87 wherein said network comprises the Internet.

89. The method of claim 87 wherein said network comprises a nodal television distribution network.

90. A method, comprising:

detecting a plurality of different types of current input patterns from use of a terminal device by a current subset of users, said input patterns including a clickstream;

performing a soft match of each of said plurality of different types of current input patterns with a plurality of stored input patterns for each of said types of input patterns, said stored patterns representing input patterns for various subsets of a set of possible users of said terminal device, said soft matches generating scored possible matches for each of said different types of data;

determining possible combinations of said scored possible matches;

determining a score for each said combination; and

identifying said current subset of users of said terminal device from said set of possible users by selecting a subset of users associated with said combination as the current subset of users for the combination having score indicating a substantial match.

91. The method of claim 90 further comprising retraining said plurality of stored input patterns with said current input patterns.

92. The method of claim 90 further comprising determining a personal profile associated with the current subset of users.

93. The method of claim 92 further comprising transmitting targeted content to said terminal device in accordance with said personal profile.

94. The method of claim 90 wherein said clickstream pattern relates to particular Web sites visited by the current subset of users or the duration of visits to the Web sites.

95. The method of claim 90 wherein said clickstream pattern relates to particular programs or channels selected by the subset of users or the duration of viewing of said programs or channels.

96. The method of claim 90 wherein said different types of current input patterns include a user keystroke pattern.

97. The method of claim 96 wherein said keystroke pattern includes digraph Interval data.

98. The method of claim 90 wherein said different types of current input patterns include user mouse usage data.

99. The method of claim 90 wherein said different types of current input patterns include user remote control usage data.

100. The method of claim 90 wherein said terminal device comprises a computer.

101. The method of claim 90 wherein said terminal device comprises a television set top box.

102. The method of claim 90 wherein said steps are implemented in a computer, and said computer communicates with said terminal device over a network.

103. The method of claim 102 wherein said network comprises the Internet.

104. The method of claim 102 wherein said network comprises a nodal television distribution network.