User information inferring system

• US 20020065836A1
• Filed: 09/18/2001
• Published: 05/30/2002
• Est. Priority Date: 09/20/2000
• Status: Active Grant

• Current Assignee: DENSO Corporation
• Original Assignee: DENSO Corporation
• Inventors: Sasaki, Mikio
• Granted Patent: US 6,957,207 B2
• Time in Patent Office: Days
• US Class Current: 707/200
• CPC Class Codes: G06F 16/21 Design, administration or maintenance of databases
• CPC Class Codes: G06F 16/24564 Applying rules; Deductive queries
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling

Abstract

A user information inferring system defines user information according to five categories, that is, user environment, circumstance, condition, request, and action, and further defines a sixth category, that is, situation data encompassing these other five categories. User information in each of the defined categories, that is, environment, circumstance, condition, request, action, and situation, is then described using item values, and operating units infer the item values. The inference operation uses temporal-spatial characteristics indicative of the effect of temporal-spatial elements on the item values, information about inferred situations, a rule database containing rules describing the relationships between items, and an optimization process for the defined categories using vector expressions.

Claims

1. A user information inferring system comprising:

inferring means for inferring user information, which is information relating to a user, as item values in at least one data set by categorically defining the user information as at least one data set and expressing the data set with items written in data lists corresponding to the data set, wherein at least a part of the item values in each data set changes in conjunction with temporal-spatial elements; and

calculating means for calculating at least one of a temporal characteristics that a temporal element imparts to the item value and a spatial characteristics that a spatial element imparts to the item value,

wherein the inferring means infers the item value based on the characteristics calculated by the characteristics calculating means.

2. The user information inferring system as in claim 1, wherein:

the characteristics calculating means calculates a periodic temporal characteristics indicative of periodicity of the temporal element.

3. The user information inferring system as in claim 1, wherein:

the characteristics calculating means calculates the temporal characteristics using a time list of temporal expressions.

4. The user information inferring system as in claim 3, wherein:

the time list describes in addition to the temporal expressions a time difference from a reference time; and

the characteristics calculating means calculates the temporal characteristics with reference to the time difference.

5. The user information inferring system as in claim 4, wherein:

the reference time is Greenwich Mean Time.

6. The user information inferring system as in claim 3, wherein:

the time list includes a time data list in which the temporal expressions or the temporal expressions and time difference are described by computer-processable variables.

7. The user information inferring system as in claim 1, wherein:

the characteristics calculating means calculates a structural space characteristics indicative of structure of the spatial element.

8. The user information inferring system as in claim 1, wherein:

the characteristics calculating means calculates the spatial characteristics using a spatial list describing a reference facility and neighboring facilities.

9. The user information inferring system as in claim 8, wherein:

the spatial list includes a location data list indicating a facility location using a vector variable from the reference facility.

10. The user information inferring system as in claim 1, further comprising:

a rule database storing rules correlating items written in the data list for the data set,

wherein the inferring means infers the item value based on the rules stored in the rule database.

11. The user information inferring system as in claim 1, further comprising:

distance calculating means,

wherein the data set is expressed by a data vector for each data set using the item value as a component,

wherein the distance calculating means calculates a distance between the data vector and a reference vector indicating an optimum condition for the data set, and

wherein the inferring means infers the item value to optimize the data set based on the distance calculated by the distance calculating means.

12. The user information inferring system as in claim 1, wherein:

the data set includes a situation data set indicating a user situation; and

the inferring means infers an item value of the situation data set based on item values of other data sets.

13. The user information inferring system as in claim 1, wherein:

each data list includes a plurality of lists hierarchically arranged according to the usage of the items; and

the inferring means preferentially infers the item value for the item written in the list with a high usage ranking.

14. The user information inferring system as in claim 1, wherein:

the inferring means infers the item value based on individual data indicative of user characteristics.

15. The user information inferring system as in claim 1, further comprising:

undetermined value inference means for inferring an incomplete or indeterminate item value from other elements after the inferring means infers the item value.

17. The user information inferring system as in claim 16, further comprising:

rule calculation means for analytically calculating rules written in symbolic notation in the rule database.

18. The user information inferring system as in claim 17, wherein:

the rule calculation means converts the symbolic notations to a specific programming language.

19. The user information inferring system as in claim 16, wherein:

the rule database stores rules for changing an item value due to the temporal and spatial elements.

20. The user information inferring system as in claim 16, wherein:

the rule database stores rules for changing an item value due to compatibility and causal relation between item values.

21. The user information inferring system as in claim 16, wherein:

the rule for changing the item value is expressed as a probability in the rule database.

22. The user information inferring system as in claim 16, wherein:

the rule database stores a rule for concretizing the item.

23. The user information inferring system as in claim 16, further comprising:

distance calculating means,

wherein the data set is expressed by a data vector for each data set using the item value as a component,

wherein the distance calculating means calculates a distance between the data vector and a reference vector indicating an optimum condition for the data set, and

wherein the inferring means infers the item value to optimize the data set based on the distance calculated by the distance calculating means.

24. The user information inferring system as in claim 16, wherein:

the data set includes a situation data set indicating a user situation; and

the inferring means infers an item value of the situation data set based on item values of other data sets.

25. The user information inferring system as in claim 16, wherein:

each data list includes a plurality of lists hierarchically arranged according to the usage of the items; and

the inferring means preferentially infers the item value for the item written in the list with a high usage ranking.

26. The user information inferring system as in claim 16, wherein:

the inferring means infers the item value based on individual data indicative of user characteristics.

27. The user information inferring system as in claim 16, further comprising:

undetermined value inference means for inferring an incomplete or indeterminate item value from other elements after the inferring means infers the item value.

29. The user information inferring system as in claim 28, wherein:

the item value of the data set is a numeric value in a specific range of which the optimum value is one of 0, a minimum value and a maximum value.

30. The user information inferring system as in claim 28, wherein:

the distance calculating means calculates the distance with consideration for the importance of each component of the vector.

31. The user information inferring system as in claim 30, wherein:

the distance calculating means calculates the distance with reference to importance of each component by giving a specific component with priority to another component.

32. The user information inferring system as in claim 30, wherein:

the distance calculating means calculates the distance with reference to importance of each component by weighting each component.

33. The user information inferring system as in claim 28, wherein:

the data set includes a situation data set indicating a user situation; and

the inferring means infers an item value of the situation data set based on item values of other data sets.

34. The user information inferring system as in claim 28, wherein:

each data list includes a plurality of lists hierarchically arranged according to the usage of the items; and

the inferring means preferentially infers the item value for the item written in the list with a high usage ranking.

35. The user information inferring system as in claim 28, wherein:

the inferring means infers the item value based on individual data indicative of user characteristics.

36. The user information inferring system as in claim 28, further comprising:

undetermined value inference means for inferring an incomplete or indeterminate item value from other elements after the inferring means infers the item value.

38. The user information inferring system as in claim 37, wherein:

the other data sets categorically define user environment, circumstance, condition, request, and action.

39. The user information inferring system as in claim 37, further comprising:

encoding means for encoding item values of the situation data set as a situation code.

40. The user information inferring system as in claim 37, wherein:

the inferring means infers item values of other data sets based on the item value of a situation data set evaluated previously.

41. The user information inferring system as in claim 37, wherein:

the inferring means decomposes an evaluated situation data set to infer an item value of another data set.

42. The user information inferring system as in claim 37, further comprising:

characteristics calculating means for calculating at least one of temporal characteristics and spatial characteristics based on the item values of a situation data set evaluated previously.

43. The user information inferring system as in claim 37, further comprising:

priority determination means for determining priority of the item value in the situation data set according to the inferred data set; and

at least one of the inferring means and the characteristics calculating means infers and calculates using the item value of the situation data set according to a priority sequence determined by the priority determination means.

44. The user information inferring system as in claim 37, wherein:

each data list includes a plurality of lists hierarchically arranged according to the usage of the items; and

the inferring means preferentially infers the item value for the item written in the list with a high usage ranking.

45. The user information inferring system as in claim 44, wherein:

the data list includes at least a complete list of all items, and an active list of most frequently used items.

46. The user information inferring system as in claim 44, wherein:

the inferring means changes the hierarchy of items in the data list based on the inferred item value.

47. The user information inferring system as in claim 37, wherein:

the inferring means infers the item value based on individual data indicative of user characteristics.

48. The user information inferring system as in claim 47, further comprising:

individual data construction means for constructing individual data.

49. The user information inferring system as in claim 48, wherein:

the individual data construction means constructs the individual data by inference based on sensor data.

50. The user information inferring system as in claim 47, wherein:

the individual data includes user preference information.

51. The user information inferring system as in claim 47, wherein:

the user preference information is determined based on user requests determined from the item value in addition to or substitutionally for the sensor data.

52. The user information inferring system as in claim 37, further comprising:

undetermined value inference means for inferring an incomplete or indeterminate item value from other elements after the inferring means infers the item value.

53. The user information inferring system as in claim 52, wherein:

the data set is expressed with a data vector,

wherein the undetermined value inference means calculates a distance to a typical data vector to infer the incomplete or indeterminate item value.

16. A user information inferring system comprising:

inferring means for inferring user information, which is information relating to a user, as an item value in at least one of data set by categorically defining the user information as at least one of data set and expressing the data set with an item written in a data list corresponding to the data set; and

a rule database storing rules correlating items written in the data list for the data set,

wherein the inferring means infers the item value based on the rules recorded in the rule database.

28. A user information inferring system comprising:

inferring means for inferring user information, which is information relating to a user, as an item value in at least one data set by categorically defining the user information as at least one data set and expressing the data set with an item written in a data list corresponding to the data set, wherein the data set is expressed by a data vector for each data set using the item value as a component; and

distance calculating means for calculating a distance between the data vector and a reference vector indicating an optimum condition for the data set,

wherein the inferring means infers the item value to optimize the data set based on the distance calculated by the distance calculating means.

37. A user information inferring system comprising:

inferring means for inferring user information, which is information relating to a user, as an item value in at least one data set by categorically defining the user information as at least one data set and expressing the data set with an item written in a data list corresponding to the data set,

wherein each data set includes a situation data set indicating a user situation, and

wherein the inferring means infers an item value of the situation data set based on item values of other data sets.