Method and Apparatus for Characterizing User Behavior Patterns from User Interaction History

• US 20120296909A1
• Filed: 02/04/2010
• Published: 11/22/2012
• Est. Priority Date: 02/04/2010
• Status: Active Grant

• Current Assignee: RPX Corporation
• Original Assignee: Nokia Corporation
• Inventors: Cao, Happia, Tian, Jilei
• Granted Patent: US 8,954,452 B2
• Time in Patent Office: Days
• US Class Current: 707/737
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

An approach is provided for characterizing user behavior patterns. The behavior pattern platform receives a plurality of context records from a device. Next, the behavior pattern platform places one or more contexts from the context records. Then, the behavior pattern platform places the contexts into one or more context groups. Then, the behavior pattern platform receives interaction data from the device, associates the context groups with the interaction data, and determines a behavior pattern based, at least in part, on the association of the context groups and the interaction data.

Claims

1-38. -38. (canceled)

39. A method comprising:

determining one or more contexts from one or more context records;

placing the one or more contexts into one or more context groups;

associating the one or more context groups with interaction data; and

determining a behavior pattern based, at least in part, on the association of the one or more context groups and the interaction data.

40. A method of claim 39, wherein each of the contexts includes a context timestamp, and wherein the placing of the contexts into the context groups comprises:

arranging the contexts according to the context timestamps, wherein the placing of the contexts into the context groups is determined based, at least in part, on a similarity between adjacently arranged contexts; and

determining a time range for each of the context groups based, at least in part, on the respective context timestamps of the contexts placed in the context groups.

41. A method of claim 40, wherein each of the context records includes one or more context features, and the method further comprises:

selecting one or more of the context features to represent the contexts;

matching the adjacently contexts based, at least in part, on the selected context features,

wherein the similarity between adjacently arranged contexts is based, at least in part, on the matching.

42. A method according to any one of claim 40, wherein each element of the interaction data includes an interaction timestamp, and the method further comprising:

matching the interaction timestamps against the context group time ranges,

wherein the associating of the context groups with the interaction data is based, at least in part, on the matching.

43. A method according to any one of claim 39, wherein the associating of the context groups with the interaction data comprises:

determining one or more context patterns from the context groups with respect to one or more elements of the interaction data;

computing a confidence for each context pattern; and

determining an association between the context pattern and the respective one or more elements of the interaction data based, at least in part, on the respective confidence.

44. A method according to any one of claim 39, further comprising:

causing, at least in part, application of the user behavior pattern to user segmentation, personalized recommendations, targeted advertising, or a combination thereof.

45. A method according to any one of claim 39, further comprising:

computing a differentiating factor for each of the context groups, the interaction data, or a combination thereof,

wherein the differentiating factor is a measure of an ability of the respective context group, interaction data, or combination thereof, to discriminate the determined behavior pattern from other behavior patterns.

46. A method according to any one of claim 39, wherein the context records include a date, a time, a location, a speed, an environmental condition, a weather condition, a profile, or a combination thereof.

47. A method according to any one of claim 39, wherein the interaction data comprises data on use applications, services, communication functions, or a combination thereof.

48. An apparatus comprising:

at least one processor; and

at least one memory including computer program code,

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus to perform at least the following,

determine one or more contexts from one or more context records;

place the one or more context records into one or more context groups;

associate the one or more context groups with interaction data; and

determine a behavior pattern based, at least in part, on the association of the one or more context groups and the interaction data.

49. An apparatus of claim 48, wherein each of the contexts includes a context record timestamp, and wherein during the placing of the contexts into the context groups the apparatus is further caused, at least in part, to:

arrange the contexts according to the context timestamps, wherein the placing of the contexts into the context groups is determined based, at least in part, on a similarity between adjacently arranged contexts; and

determine a time range for each of the context groups based, at least in part, on the respective context timestamps of the contexts placed in the context groups.

50. An apparatus of claim 49, wherein each of the context records includes one or more context features, and the apparatus is further caused, at least in part, to:

select one or more of the context features to represent the contexts;

match the adjacently contexts based, at least in part, on the selected context features,

wherein the similarity between adjacently arranged contexts is based, at least in part, on the matching.

51. An apparatus according to any one of claim 49, wherein each element of the interaction data includes an interaction timestamp, and the apparatus is further caused, at least in part, to:

match the interaction timestamps against the context group time ranges,

wherein the associating of the context groups with the interaction data is based, at least in part, on the matching.

52. An apparatus according to any one of claim 48, wherein during the associating of the context groups with the interaction data the apparatus is further caused, at least in part, to:

determine one or more context patterns from the context groups with respect to one or more elements of the interaction data;

compute a confidence for each context pattern; and

determine an association between the context pattern and the respective one or more elements of the interaction data based, at least in part, on the respective confidence.

53. An apparatus according to any one of claim 48, wherein the apparatus is further caused, at least in part, to:

cause, at least in part, application of the user behavior pattern to user segmentation, personalized recommendations, targeted advertising, or a combination thereof.

54. An apparatus according to any one of claim 48, wherein the apparatus is further caused, at least in part, to:

computing a differentiating factor for each of the context groups, the interaction data, or a combination thereof,

wherein the differentiating factor is a measure of an ability of the respective context group, interaction data, or combination thereof, to discriminate the determined behavior pattern from other behavior patterns.

55. An apparatus according to any one of claim 48, wherein the context records include a date, a time, a location, a speed, an environmental condition, a weather condition, a profile, or a combination thereof.

56. An apparatus according to any one of claim 48, wherein the interaction data comprises data on use applications, services, communication functions, or a combination thereof.

57. An apparatus according to any one of claim 48, wherein the apparatus is a mobile phone further comprising:

user interface circuitry and user interface software configured to facilitate user control of at least some functions of the mobile phone through use of a display and configured to respond to user input; and

a display and display circuitry configured to display at least a portion of a user interface of the mobile phone, the display and display circuitry configured to facilitate user control of at least some functions of the mobile phone.

58. A computer-readable storage medium carrying one or more sequences of one or more instructions which, when executed by one or more processors, cause an apparatus to at least perform the following steps:

determining one or more contexts from one or more context records;

placing the one or more context records into one or more context groups;

associating the one or more context groups with interaction data; and

determining a behavior pattern based, at least in part, on the association of the one or more context groups and the interaction data.