Smart agent based on habit, statistical inference and psycho-demographic profiling

• US 7,003,792 B1
• Filed: 11/30/1999
• Issued: 02/21/2006
• Est. Priority Date: 11/30/1998
• Status: Expired due to Term

• Current Assignee: Rovi Guides, Inc. (Adeia Inc.)
• Original Assignee: Index Systems Incorporated (Adeia Inc.)
• Inventors: Yuen, Henry C.
• Primary Examiner(s): Srivastava, Vivek
• Application Number:
  US09/449,887
• Time in Patent Office: 2,275 Days
• Field of Search: 725/39, 725/46, 725/44, 725/45, 725/47, 725/9
• US Class Current: 725/46
• CPC Class Codes: H04N 21/42204 User interfaces specially adapted for controlling a client device through a remote control device; Remote control devices therefor constructive details of casings for the remote control device H01H9/0235; user interfaces for controlling a tuning device of a television receiver through a remote control H03J9/00; remote control of peripheral devices connected to a television receiver through the remote control device of the television receiver H04B1/205
• CPC Class Codes: H04N 21/4316 for displaying supplemental content in a region of the screen, e.g. an advertisement in a separate window
• CPC Class Codes: H04N 21/44224 Monitoring of user activity on external systems, e.g. Internet browsing
• 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 21/47 End-user applications
• CPC Class Codes: H04N 21/478 Supplemental services, e.g. displaying phone caller identification, shopping application
• 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 21/812 involving advertisement data advertising per se G06Q30/02
• CPC Class Codes: H04N 5/45 Picture in picture , e.g. displaying simultaneously another television channel in a region of the screen
• CPC Class Codes: H04N 5/782 on tape
• CPC Class Codes: H04N 7/163 by receiver means only

Abstract

A smart agent (SA) is provided which resides locally in a local client device and by iterative means integrates the habit, statistics and psycho-demographic information of a user to infer the user'"'"'s preferences. The SA may also utilize the preferences to filter information delivered to the local device. The invention further discloses a means of collecting, combining, integrating and inferring information from the user to arrive at a psycho-demographic profile of the user, and a means of utilizing such psycho-demographic profile to select or filter information delivered to the user, thereby achieving targeting. The invention also discloses a means of classifying and identifying the information delivered so that it can be matched, filtered or selected.

Claims

1. A method of determining a profile of a user based on statistical data and psycho-demographic information about the user, the method comprising the steps of:

storing statistical data and psycho-demographic information that is based on probability-based character traits about a plurality of contents;

collecting data about user selections of a portion of the plurality of contents;

iteratively integrating the collected user selections data with the stored statistical data and the psycho-demographic information about the user selections of a portion of the plurality of contents; and

inferring the profile of the user from the integrated information.

9. The method of claim 1 wherein, the user is a television viewer having a television receiver and the step of collecting comprises collecting data about the viewer'"'"'s television program selections.

10. The method of claim 9 wherein, the step of collecting comprises collecting data about television programs watched by the viewer.

11. The method of claim 9 wherein, the step of collecting comprises collecting data about television programs selected in an EPG.

12. The method of claim 9 wherein, the step of collecting comprises collecting data about television programs scheduled for recording or watching in an EPG.

13. The method of claim 9 wherein, the step of inferring the profile (UP) is determined by the equation:

UP(X)={ε

1, ε

2, ε

3, . . . , ε

N)

where ε

i={(1/n) Σ

j Ei,j τ

j ω

j, . . . }, i identifies the elements, j identifies the television program being watched by the television viewer X, Ei,j is the jth television program of the ith element, providing information on the ith element, ij is the duration for which the television program j is being watched, and ω

j is a weighting function to give weight or take away weight for a given television program and having a range from 0 to 1.

14. The method of claim 9 further comprising the step of targeting information to the television viewer based on the profile.

15. The method of claim 14 wherein, the information is advertisement.

16. The method of claim 15 wherein, the advertisement includes one or more of text, still image, and video.

17. The method of claim 14 wherein, the information is displayed simultaneously with an EPG on a display.

18. The method of claim 1, further comprising the step of maintaining the profile in a secure file.

19. The method of claim 9, further comprising the step of storing an EPG data base at the receiver, the EPG data base including time, channel, and program category identifiers of telecast television programs;

and collecting step collects data about the program categories of telecast television programs selected by the viewer.

20. The method of claim 1 wherein, the user is an Internet user having an Internet terminal for accessing the Internet and the step of collecting comprises collecting data about the user'"'"'s website selections.

21. The method of claim 20 wherein, the step of collecting comprises collecting data about websites visited by the user.

22. The method of claim 21 wherein, the step of collecting comprises collecting data about products purchased the user from the visited website.

23. The method of claim 20 wherein, the accessing the Internet is initiated from an EPG.

24. The method of claim 20 wherein, the step of inferring the profile (UP) is determined by the equation:

UP(X)={ε

1, ε

2, ε

3, . . . , ε

N)

where ε

i={(1/n) Σ

j Ei,j τ

j ω

j, . . . }, i identifies the elements, j identifies the website being visited by the Internet user X, Ei,j is the jth visited website of the ith element, providing information on the ith element, τ

j is the duration for which the website j is being visited, and ω

j is a weighting function to give weight or take away weight for a given website and having a range from 0 to 1.

25. The method of claim 20 further comprising the step of targeting information to the Internet user based on the profile.

26. The method of claim 20 wherein, the information is advertisement.

27. The method of claim 26 wherein, the advertisement includes one or more of text, still image, and video.

28. The method of claim 20 wherein, the information is displayed simultaneously with an EPG on the Internet terminal.

2. A method for determining a profile of a user based on statistical data and psycho-demographic information about the user, the method comprising the steps of:

collecting data about user selections;

iteratively integrating the collected selections data with the statistical data and the psycho-demographic information about the user; and

inferring the profile of the user from the integrated information wherein, the step of inferring the profile (UP) is determined by the equation;

UP(X)={ρ

[Ei(X)]}

where Ei(x) is the ith element of the character trait of a real person X, and ρ

[Ei(X)] is the probability of the person X having a specific character trait.

3. The method of claim 2 wherein, the element Ei is a binary element.

4. The method of claim 2 wherein, the element Ei is a multiple-outcome-element.

5. The method of claim 2 wherein, the step of iteratively integrating comprises iteratively integrating the collected selections with the statistical data and the psycho-demographic information in a cumulative fashion.

6. The method of claim 5 wherein, the collected selections data is disregarded after the integration step.

7. The method of claim 5 wherein, the step of inferring the profile (UP) is determined by the equation:

UP(X)={ε

1, ε

2, ε

3, . . . , ε

N)

where υ

i={(1/n) Σ

j Ei,j τ

j ω

j, . . . }, i identifies the elements, j identifies events being utilized by the person X, Ei,j is the jth event of the ith element, providing information on the ith element, τ

j is the duration for which event j is being utilized, and ω

j is a weighting function to give weight or take away weight for a given event and having a range from 0 to 1.

8. The method of claim 7 wherein, ω

j is a function of one or more of the intensity of the event, the length of the event, the day of the week, and the time of the day.

29. A method of inferring a preference profile of a user utilizing general populous statistical data and psycho-demographic information about the user, the profile having a plurality of unique characteristic segments, the method comprising the steps of:

monitoring user selections

iteratively collecting data about the user selections with respect to each of the unique characteristic segments;

iteratively assigning probabilities to each of the collected unique characteristic segments;

integrating each of the collected unique characteristic segments with assigned probabilities derived from the general populous statistical data and with the psycho-demographic information about the user that is based on probability-based character traits; and

inferring the profile of the user from the integrated information.

35. The method of claim 29 further comprising the steps of:

recording a predetermine number of inconsistencies of the user selections;

separating the user profile into two consistent sets, a first user profile and a second user profile, while continuing collecting data for the user profile;

monitoring the time-of-day for each collected data for the first and second user profiles;

inferring that the first user profile and the second user profile are for different individuals if over a predetermined period the time-of-day of the first user profile data is grouped together and the time-of-day of the second user profile data is grouped together.

36. The method of claim 29 further comprising the step of integrating external data with the collected data.

37. The method of claim 36 wherein, the external data is one or more of user provided information, feedback information, inherent information, and inferred information.

38. The method of claim 37 wherein, the user provided information includes user assignable weighting.

30. A method for inferring a preference profile of a user utilizing general populous statistical data and psycho-demographic information about the user, the profile having a plurality of unique characteristic segments, the method comprising the steps of:

monitoring user selections

iteratively collecting data about the user selections with respect to each of the unique characteristic segments;

iteratively assigning probabilities to each of the collected unique characteristic segments;

integrating each of the collected unique characteristic segments with assigned probabilities derived from the general populous statistical data and with the psycho-demographic information about the user; and

inferring the profile of the user from the integrated information wherein, the step of inferring the profile (UP) is determined by the equation;

UP(X)={ρ

[Ei(X)]}

where Ei(x) is the ith element of the character trait of a real person X, and ρ

[Ei(X)] is the probability of the person X having a specific character trait.

31. The method of claim 30 wherein, the step of inferring the profile (UP) is determined by the equation:

UP(X)={ε

1, ε

2, ε

3, . . . , ε

N)

where ε

i={(1/n) Σ

j Ei,j τ

j ω

j, . . . }, i identifies the elements, j identifies events being utilized by the person X, Ei,j is the jth event of the ith element, providing information on the ith element, τ

j is the duration for which event j is being utilized, and ω

j is a weighting function to give weight or take away weight for a given event and having a range from 0 to 1.

32. The method of claim 31 wherein, ω

i is defined as v/N, where v is the number selections present as alternatives when a selection is made, and N is the total number of selections available.

33. The method of claim 31 wherein, the user is a television viewer having a television receiver, the events are television programs being viewed by the viewer, and the step of collecting comprises collecting data about the viewer'"'"'s television program selections.

34. The method of claim 31 wherein, the user is an Internet user having an Internet terminal for accessing the Internet, the events are websites being visited by the Internet user, and the step of collecting comprises collecting data about the user'"'"'s website selections.

39. A system for inferring preference profile of a user utilizing general populous statistical data and psycho-demographic information about the user comprising:

a plurality of unique characteristic segments included in the profile;

means for monitoring user selections;

means for iteratively collecting data about the user selections with respect to each of the unique characteristic segments;

means for iteratively assigning probabilities to each of the collected unique characteristic segments;

means for integrating each of the collected unique characteristic segments with assigned probabilities derived from the general populous statistical data and with the psycho-demographic information about the user that is based on probability-based character traits; and

means for inferring the profile of the user from the integrated information.

40. The system of claim 39 wherein, the means for integrating and the means for inferring is local to the user.

41. The system of claim 39 wherein, the means for integrating and the means for inferring is remote to the user.

42. A system for inferring preference profile of a user utilizing general populous statistical data and psycho-demographic information about the user comprising:

a plurality of unique characteristic segments included in the profile;

means for monitoring user selections;

means for iteratively collecting data about the user selections with respect to each of the unique characteristic segments;

means for iteratively assigning probabilities to each of the collected unique characteristic segments;

means for integrating each of the collected unique characteristic segments with assigned probabilities derived from the general populous statistical data and with the psycho-demographic information about the user; and

means for inferring the profile of the user from the integrated information wherein, means for inferring the profile (UP) is a microprocessor computing the equation;

UP(X)={ρ

[Ei(X)]}

where Ei(x) is the ith element of the character trait of a real person X, and ρ

[Ei(X)] is the probability of the person X having a specific character trait.

43. The system of claim 42 wherein, the means for inferring the profile (UP) is a microprocessor computing the equation:

UP(X)={ε

1, ε

2, ε

3, . . . ε

N)

where ε

i={(l/n) Σ

j Ei,j τ

j Ω

j, . . . }, i identifies the elements, j identifies events being utilized by the person X, Ei,j is the jth event of the ith element, providing information on the ith element, τ

j is the duration for which event j is being utilized, and ω

j is a weighting function to give weight or take away weight for a given event and having a range from 0 to 1.

44. The system of claim 43 wherein, the user is a television viewer having a television receiver, the events are television programs being viewed by the viewer, and the means for collecting comprises means for collecting data about the viewer'"'"'s television program selections.

45. The system of claim 43 wherein, the means for collecting comprises means for collecting data about television programs selected in an EPG.

46. The system of claim 43 wherein, the user is an Internet user having an Internet terminal for accessing the Internet, the events are websites being visited by the Internet user, and the means for collecting comprises means for collecting data about the user'"'"'s website selections.