Method and business process to maintain privacy in distributed recommendation systems

• US 6,968,334 B2
• Filed: 05/15/2001
• Issued: 11/22/2005
• Est. Priority Date: 05/15/2001
• Status: Expired due to Term

• Current Assignee: Nokia Technologies Oy (Nokia Corporation)
• Original Assignee: Nokia Corporation
• Inventors: Sorvari, Antti, Salmenkaita, Jukka-Pekka
• Primary Examiner(s): Rones, Charles
• Assistant Examiner(s): ABEL JALIL, NEVEEN
• Application Number:
  US09/854,635
• Publication Number:
  US 20020188589A1
• Time in Patent Office: 1,652 Days
• Field of Search: 707/2, 707/6, 707/10, 707/1, 707/102, 707/3, 707/104.1, 345/700, 709/210, 709/203, 709/217, 709/225, 455/11.1, 455/756, 455/714
• US Class Current: 1/1
• CPC Class Codes: G01C 21/20 Instruments for performing navigational calculations G01C21/24, G01C21/26 take precedence
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9537 Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
• CPC Class Codes: G06F 16/9577 Optimising the visualization of content, e.g. distillation of HTML documents
• CPC Class Codes: G06F 21/51 at application loading time, e.g. accepting, rejecting, starting or inhibiting executable software based on integrity or source reliability
• CPC Class Codes: G06F 21/52 during program execution, e.g. stack integrity ; Preventing unwanted data erasure; Buffer overflow
• CPC Class Codes: G06F 21/6245 Protecting personal data, e.g. for financial or medical purposes
• CPC Class Codes: H04L 63/107 wherein the security policies are location-dependent, e.g. entities privileges depend on current location or allowing specific operations only from locally connected terminals
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/04 specially adapted for terminals or networks with limited capabilities; specially adapted for terminal portability
• CPC Class Codes: H04L 67/12 specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
• CPC Class Codes: H04L 67/52 specially adapted for the location of the user terminal
• CPC Class Codes: H04L 67/56 Provisioning of proxy services store-and-forward switching systems in data switching networks H04L12/54
• CPC Class Codes: H04L 67/564 Enhancement of application control based on intercepted application data
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
• CPC Class Codes: H04M 7/0024 Services and arrangements where telephone services are combined with data services arrangements which combine a telephonic equipment and a computer H04M7/0012
• CPC Class Codes: H04W 4/02 Services making use of location information
• CPC Class Codes: H04W 4/029 Location-based management or tracking services
• CPC Class Codes: H04W 4/18 Information format or content conversion, e.g. adaptation by the network of the transmitted or received information for the purpose of wireless delivery to users or terminals
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99934 : Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99935 : Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 : Pattern matching access

Abstract

A distributed recommendation system and method are disclosed that provides greater privacy for the user'"'"'s private data. The method distributes the tasks of a recommendation system between wireless devices and network servers, so as to protect the privacy of end users. The system and method provide greater privacy for context-sensitive, adaptive, user interfaces for Internet service usage by wireless devices. The users wireless device sends a current context-activity pair to a network server in response to either the user'"'"'s selection of an activity or automatically. The server then responds an appropriate recommendation. Recommendations received from the server by the wireless device are filtered to identify new or significant information. As the system makes new recommendations to users in response to context-activity pairs submitted by their wireless devices, the server gathers the new context-activity pairs and recommendations and adds them to its context-activity pair database.

Claims

1. A wireless device to provide recommendations to a user that are appropriate to the device'"'"'s current environment, comprising:

a sensor for providing sensor signals as a metadata vector which represents the current sensor signals and characterizes the current state of the wireless device

characterizing a current environment of the wireless device;

a context inference engine coupled to the sensor, for processing the sensor signals said context inference engine providing a current context result from the processing;

a processor coupled to the context inference engine, for forming a context-activity pair by selecting an activity and pairing with the current context result;

a database coupled to the processor, for providing recommendations using the context-activity pair without user identification wherein the database comprises a table listing context-activity pairs each related to (i) a listing of service recommendations and (ii) a listing of number times recommended for each service recommendation; and

an output device coupled to the database, for providing the recommendations to the user in response to the context-activity pair.

2. The wireless device of claim 1 further comprising:

means for matching contact-activity pairs in the database similar to the pair received from the context inference engine; and

the output device providing alternative recommendations to the user in response to the context-activity pair.

3. A system to provide recommendations to a user of a wireless device that are appropriate to the device'"'"'s current environment, comprising:

a sensor in the wireless device for providing sensor signals as a metadata vector representing a current sensor signals and characterizing a current environment of the wireless device;

appending a message authentication code and digital signature to insure the integrity of the metadata vector;

a processor coupled to the sensor, for forming pair information by selecting an activity and pairing the activity with current sensor information derived from said sensor signals, said processor sending the pair information to a server;

a context inference engine in the server coupled to the wireless device, for processing the context-activity pair information, said context inference engine providing a current context result from the processing;

a database coupled to the context inference engine, for providing recommendations using the activity and current result without user identification wherein the database comprises a table listing context-activity pairs each related to (i) a listing of service recommendations and (ii) a listing of number times recommended for each service recommendation; and

an output device in the wireless device and coupled to the database, for providing recommendations to the user.

4. The system of claim 3 further comprises

the database includes context-activity pairs and related service recommendations;

a privacy control block controlling access of applications to private context information; and

the output device provides alternative recommendations to the user.

5. An apparatus to enable a wireless device to provide a recommendations to a user that are appropriate to the device'"'"'s current environment, comprising:

a processor;

a memory coupled to the processor, programmed to perform the steps of;

receiving sensor signals characterizing a current environment of the wireless device;

processing the sensor signals with a context inference engine;

outputting a current context result from the processing by the context inference engine;

forming a context-activity pair by selecting an activity and pairing with the current context result;

causing a database of recommendations to be searched using the context-activity pair without user identification wherein the database of recommendations comprises a table listing context-activity pairs each related to (i) a listing of service recommendations and (ii) a listing of number times recommended for each service recommendation; and

providing recommendations to the user in response to the searching step.

6. The apparatus of claim 5 wherein the memory steps further comprise:

forming a database of context-activity pairs and related service recommendations in a remote server;

controlling access of applications to private context information via a privacy control block; and

providing alternative recommendations to the user for the selection of a context-activity using a recommendation algorithm in response to the searching step.

7. The apparatus of claim 5, wherein the processing of the sensor signals with a context inference engine is embodied as programmed instructions executed within the users'"'"'s wireless device.

8. The apparatus of claim 5, wherein the processing of the sensor signals with a context inference engine is embodied as programmed instructions executed within a separate network server in response to signals from the user'"'"'s wireless device.

9. A method to enable a wireless device to provide recommendations to a user that are appropriate to the device'"'"'s current environment, comprising:

receiving sensor signals as a metadata vector characterizing a current environment of the wireless device;

appending a message authentication code and digital signature to insure the integrity of the metadata vector;

processing the sensor signals with a context inference engine to produce a set of current context results;

forming a set of context-activity pairs by selecting an activity and pairing with the set of current context results;

searching a set of related service history items from a history log using the set of context-activity pairs;

forming context-activity pair information from the set of context-activity pair and the set of related service history items;

searching a database of recommendations using the context-activity pair information without user identification wherein the database of recommendations comprises a table listing context-activity pairs each related to (i) a listing of service recommendations and (ii) a listing of number times recommended for each service recommendation; and

providing recommendations to the user in response to the searching step.

10. A method to enable a wireless device to provide recommendations to a user that are appropriate to the device'"'"'s current environment, comprising:

receiving sensor signals as a metadata vector characterizing a current environment of the wireless device;

processing the sensor signals with a context inference engine utilizing adaptive and continuous learning processes;

outputting a current context result from the sensor processing by the context inference engine;

selecting a user activity and pairing the activity with the current context result to form a context-activity pair;

searching a database of context-activity pairs in a recommendation system while maintaining the privacy of the user wherein the database comprises a table listing context activity pairs each related to (i) a listing of service recommendations and (ii) a listing of number times recommended for each service recommendation;

providing recommendations to the user relative to the context-activity pair in response to the searching step;

filtering the recommendations using an algorithm to identify new and significant information as new recommendations; and

displaying the new recommendations to the user.

11. The method of claim 10 further comprising:

forming a metadata vector of the sensor signals for processing in the device or transmission to a server for processing.

12. The method of claim 10 further comprising:

establishing a privacy user interface to a privacy control element enabling the user to set privacy policies related to access to the context inference engine, and to alert the user that an application program is attempting to register to receive the user'"'"'s private context awareness information.

13. The method of claim 10 further comprising:

excluding user personal data from the database of context-activity pairs.

14. The method of claim 10, further comprising:

providing context activity pair data sets in the database to third parties for market research.

15. The method of claim 10, further comprising:

forming a database of context-activity pairs and related service recommendations;

controlling access of applications to private context information via a privacy control block; and

wherein

alternative recommendations are provided to the user for the selection of a context activity using a recommendation algorithm.

16. A method to enable a wireless device to provide recommendations to a user that are appropriate to the device'"'"'s current environment, comprising:

receiving sensor signals characterizing a current environment of the wireless device;

processing the sensor signals with a context inference engine to produce a set of current context results;

forming a set of context-activity pairs by selecting an activity and pairing with the set of current context results;

accessing a set of related service history items from a history log;

forming context-activity pair information from the set of current context results and the set of related service history items;

searching a database of recommendations using the context-activity pair without user identification wherein the database of recommendations comprises a table listing context-activity pairs each related to (i) a listing of service recommendations and (ii) a listing of number times recommended for each service recommendation; and

providing recommendations to the user in response to the searching step.

17. The method of claim 16, which further comprises:

providing the recommendation in a separate wireless network server in response to context-activity pair information received at the server from the user'"'"'s wireless device.

18. The method of claim 17, which further comprises:

maintaining the database as a context-activity pair database by the separate wireless network server;

associating in the database the context-activity pair information with appropriate recommendations made in the past to many users.

19. The method of claim 18, which further comprises:

making new recommendations to the user in response to the context-activity pair information submitted by the wireless device; and

gathering the new recommendations and adding them to the database;

whereby the variety, quality and pertinence of the recommendations in the database grows as the recommendation system is used.

20. The method of claim 19, which further comprises:

compiling statistical usage information about the recommendations and storing the usage information in the database.

21. The method of claim 20, which further comprises:

providing the statistical usage information to the wireless device accompanying the recommendations.

22. The method of claim 21, which further comprises:

filtering the recommendations received at the wireless device by using the statistical usage information accompanying the recommendations.

23. The method of claim 22, wherein said providing step further comprises:

filtering the recommendations at the wireless device using statistical usage information associated with the recommendations.

24. The method of claim 23, wherein said providing step further comprises:

accessing a history log of previous recommendations provided to the user;

filtering new recommendations from the previous recommendations and providing the new recommendations to the user.

25. The method of claim 24, wherein said providing step further comprises:

accessing a history log of previous recommendations provided to the user, including ratings of the previous recommendations;

filtering recommendations using the ratings and providing the filtered recommendations to the user.

26. The method of claim 16 further comprising:

forming a database of context-activity pairs and related service recommendations;

controlling access of applications to private context information via a privacy control block;

matching contact-activity pairs in the database similar to the pair received from the context inference engine; and

wherein

alternative recommendations are provided to the wireless device from the database for the selection of a context- activity using a recommendation algorithm.

27. A business method to enable a wireless device to provide recommendations to a user that are appropriate to the device'"'"'s current environment, comprising:

representing with a metadata vector the current sensor signals characterizing a current environment of the wireless device with a current context result;

processing of the sensor signals with a context inference engine embodied as programmed instructions executed within a separate wireless network server in response to signals from the user'"'"'s wireless device;

appending a message authentication code and digital signature to insure the integrity of the metadata vector;

forming a context-activity pair by selecting an activity and pairing with the current context result;

accessing a database of recommendations using the context-activity pair, the database of recommendations excluding any user personal data;

the database coupled to the processor, for providing recommendations using the context-activity pair without user identification wherein the database comprises a table listing context-activity pairs each related to (i) a listing of service recommendations and (ii) a listing of number times recommended for each service recommendation; and

providing recommendations to the wireless device from the database.

28. The business method of claim 27, which further comprises:

gathering new recommendations and adding them to the database without including any user personal data or old recommendations.

29. The business method of claim 28, which further comprises:

providing at least portions of the database to a third party service provider.

30. The business method of claim 27, which further comprises:

compiling statistical usage information about the recommendations and storing the usage information in the database.

31. The business method of claim 30, which further comprises:

providing the statistical usage information to the wireless device accompanying the recommendations.

32. The business method of claim 27, which further comprises:

filtering the recommendations received at the wireless device by using the statistical usage information accompanying the recommendations.

33. The business method of claim 27, which further comprises:

accessing a history log of previous recommendations provided to the user;

filtering new recommendations from the previous recommendations and providing the new recommendations to the user.

34. The business method of claim 27, which further comprises:

accessing a history log of previous recommendations provided to the user, including ratings of the previous recommendations,

filtering recommendations using the ratings and providing the filtered recommendations to the user.

35. The business method of claim 27, which further comprises:

providing the recommendations to an application program layer.

36. The business method of claim 27 further comprising:

forming a database of context-activity pairs and related service recommendations;

controlling access of applications to private context information via a privacy control block; and

wherein

alternative recommendations are provided to the wireless device from the database for the selection of a context-activity using a recommendation algorithm.

37. The method of claim 36 wherein at least one source of services matching the context-activity pair received from the wireless device.

38. A method to enable a wireless device to provide recommendations to a user that is appropriate to the device'"'"'s current environment, comprising:

receiving sensor signals characterizing a current environment of the wireless device;

processing the sensor signals with a context inference engine;

outputting a current context result from the processing by the context inference engine;

forming a context-activity pair by selecting an activity and pairing with the current context result;

searching a database of recommendations using the context-activity pair without user identification wherein the database of recommendations comprises a table listing context-activity pairs each related to (i) a listing of service recommendations and (ii) a listing of number times recommended for each service recommendation; and

providing recommendations to the user in response to the searching step.

39. The method of claim 38, wherein the processing of the sensor signals with a context inference engine is embodied as programmed instructions executed within the user'"'"'s wireless device.

40. The method of claim 38, wherein the processing of the sensor signals with a context inference engine is embodied as programmed instructions executed within a separate network server in response to signals from the user'"'"'s wireless device.

41. The method of claim 40, which further comprises:

providing to the user control over the privacy of the user'"'"'s information within the network server.

42. The method of claim 41, which further comprises:

maintaining the database as a context-activity pair database by the server, which contains no personal information about the user;

associating in the database the context-activity pair information with appropriate recommendations made in the past to many users.

43. The method of claim 42, which further comprises:

making new recommendations to the user in response to the context-activity pair information submitted by the wireless device; and

gathering the new recommendations and adding them to the database without any personal information about the user.

44. The method of claim 40, wherein the signals from the user'"'"'s wireless device are sent to the server without any user identification.

45. The method of claim 40, wherein the wireless device offloads a portion of the processing of the sensor signals to a context inference engine to the server.

46. The method of claim 38, wherein the sensor signals are selected from the group consisting of positioning signals, touch signals, audio signals, compass signals, ambient light signals, ambient temperature signals, three-axis acceleration signals, time signals, and the device'"'"'s operational mode signals.

47. The method of claim 38, wherein the selecting of an activity is automatically performed in the wireless device.

48. The method of claim 38, wherein the selecting of an activity performed by the user in the wireless device.

49. The method of claim 38, which further comprises:

providing the recommendation in a separate server in response to context-activity pair information received at the server from the user'"'"'s wireless device.

50. The method of claim 49, which further comprises:

maintaining the database as a context-activity pair database by the server;

associating in the database the context-activity pair information with appropriate recommendations made in the past to many users.

51. The method of claim 50, which further comprises:

making new recommendations to the user in response to the context-activity pair information submitted by the wireless device; and

gathering the new recommendations and adding them to the database;

whereby the variety, quality and pertinence of the recommendations in the database grows as the recommendation system is used.

52. The method of claim 51, which further comprises:

compiling statistical usage information about the recommendations and storing the usage information in the database.

53. The method of claim 52, which further comprises:

providing the statistical usage information to the wireless device accompanying the recommendations.

54. The method of claim 53, which further comprises:

filtering the recommendations received at the wireless device by using the statistical usage information accompanying the recommendations.

55. The method of claim 51 wherein new recommendations generated by the server are added to the database for statistical purposes and expanded user selections.

56. The method of claim 38, wherein said providing step further comprises:

filtering the recommendations at the wireless device using statistical usage information associated with the recommendations.

57. The method of claim 38, wherein said providing step further comprises:

accessing a history log of previous recommendations provided to the user;

filtering new recommendations from the previous recommendations and providing the new recommendations to the user.

58. The method of claim 38, wherein said providing step further comprises:

accessing a history log of previous recommendations provided to the user, including ratings of the previous recommendations;

filtering recommendations using the ratings and providing the filtered recommendations to the user.

59. The method of claim 38, which further comprises:

providing the recommendations to an application program.

60. The method of claim 59 wherein the application program receives recommendations from application program interfaces for further processing without showing the recommendation to the user.

61. The method of claim 38 further comprising:

forming a database of context-activity pairs and related service recommendations in a remote server;

controlling access of applications to private context information via a privacy control block;

matching contact-activity pairs in the database similar to the pair received from the context inference engine; and

providing alternative recommendations to the user for the selection of a context-activity using a recommendation algorithm in response to the searching step.

62. The method of claim 38 wherein a metadata vector represents the current sensor signals;

characterizes the current state of the wireless device, and combined with the user activity as the context-activity pair.