MOBILE DEVICE CONTROLLED BY CONTEXT AWARENESS

• US 20120271913A1
• Filed: 11/09/2010
• Published: 10/25/2012
• Est. Priority Date: 11/11/2009
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Adfore Technologies Oy
• Inventors: Tallgren, Markus, Niemi, Joni
• Granted Patent: US 8,990,384 B2
• Time in Patent Office: Days
• US Class Current: 709/217
• CPC Class Codes: H04W 24/00 Supervisory, monitoring or testing arrangements
• 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 4/50 Service provisioning or reconfiguring
• CPC Class Codes: H04W 88/02 Terminal devices

Abstract

A method for controlling a mobile device on the basis of context awareness includes: recognizing (101) changes of context related to the mobile device and/or a user of the mobile device, gathering (102) usage data indicative of control actions given by the user and directed to the mobile device during different recognized changes of context, selecting (103), as a response to a change of context, at least one service action from among a pre-determined set of service actions related to services provided with the mobile device at least partly on the basis of the usage data, and controlling (104) the mobile device to perform the selected at least one service action, e.g. adaptation of a user interface of the mobile device. As a consequence of gathering the usage data, the operation of the mobile device in different changes of context can be tailored for the user.

Claims

1-44. -44. (canceled)

45. A mobile device comprising a processing system arranged to:

recognize a change of context related to the mobile device and/or a user of the mobile device,

control the mobile device to perform at least one service action related to services provided with the mobile device as a response to a recognized change of context,

gather usage data indicative of control actions given by the user and directed to the mobile device during different recognized changes of context, and

select, prior to performing the at least one service action, the at least one service action from among a pre-determined set of service actions at least partly on the basis of the usage data.

46. A mobile device according to claim 45, wherein the processing system is arranged to use, in addition to the usage data for the selection of the service action, history data indicative of control actions given by users of other mobile devices during different changes of context related to the other mobile devices and/or the users of the other mobile devices.

47. A mobile device according to claim 45, wherein the processing system is arranged analyze, with a data mining algorithm, correlations between different changes of context and different service actions from a data base comprising the usage data.

48. A mobile device according to claim 46, wherein the processing system is arranged analyze, with a data mining algorithm, correlations between different changes of context and different service actions from a data base comprising the usage data and the history data.

49. A mobile device according to claim 46, comprising a data interface for downloading the history data from an external device.

50. A mobile device according to claim 46, comprising a receiver for receiving the history data via a mobile communication network.

51. A mobile device according to claim 45, comprising one or more sensing entities for obtaining sensing data indicative of the context related to the mobile device and/or the user of the mobile device, the processing system being arranged to recognize the change of context on the basis of the sensing data.

52. A mobile device according to claim 45, comprising a data interface for receiving, from one or more external device, sensing data indicative of the context related to the mobile device and/or the user of the mobile device, the processing system being arranged to recognize the change of context on the basis of the sensing data.

53. A mobile device according to claim 51, wherein the processing system is arranged, for recognizing the change of context, to:

determine a vector of feature values on the basis of the sensing data, and

classify the vector of feature values to one of alternative context classes, the change of context being present when the context class determined by the classification of the vector of feature values differs from a previously determined context class.

54. A mobile device according to claim 53, wherein the processing system is arranged to adapt an algorithm used in the classification on the basis of the vector of feature values and feedback information given by the user.

55. A mobile device according to claim 54, wherein the processing system is arranged to adapt the algorithm so that an ideal feature vector of the context class determined by the classification is moved, depending on the feedback information, closer to or farther away from the vector of feature values, the ideal feature vector of each context class being such a vector of feature values that best matches with that particular context class and the context class determined by the classification being the one whose ideal feature vector is closest to the vector of feature values in the sense of a pre-determined vector norm.

56. A mobile device according to claim 53, wherein the processing system is arranged to:

check whether classification of the vector of feature values would fulfill a pre-determined quality criterion,

check availability of data transfer connections to and from a server device,

as a response to a situation in which the data transfer connections are available and the classification would not fulfill the pre-determined quality criterion, transmit the vector of feature values to the server device in order to enable the server device to classify the vector of feature values to one of the alternative context classes, and

as a response to receiving a classification result from the server device, use the classification result received from the server device for determining whether the change of context has taken place.

57. A mobile device according to claim 56, wherein the processing system is arranged to calculate the minimum of distances between the vector of feature values and ideal feature vectors of the alternative context classes in order to check whether the classification of the vector of feature values would fulfill the pre-determined quality criterion, the distances being defined in the sense of a pre-determined vector norm and the ideal feature vector of each context class being such a vector of feature values that best matches with that particular context class.

58. A mobile device according to claim 57, wherein the processing system is arranged to transmit to the server device the ideal feature vector of one or more of the context classes in order to enable the server device to classify the vector of feature values to one of the alternative context classes in a same manner as the classification is carried out by the mobile device.

59. A mobile device according to claim 53, wherein the processing system is arranged to recognize the change of context on the basis of whether a context class given by the user differs from the previously determined context class.

60. A mobile device according to claim 53, wherein the processing system is arranged to:

check availability of the data transfer connections to and from the server device,

as a response to a situation in which the data transfer connections are available, transmit information based on the sensing data to a server device in order to enable the server device to form at least one second vector of feature values, and

as a response to receiving the at least one second vector of feature values from the server device, use each second vector of feature values for determining whether the change of context has taken place.

61. A mobile device according to claim 60, wherein the processing system is arranged to select one of the context classes determined by:

the classification of the vector of feature values, and

the classification of each of the at least one second vectors of feature values,

the change of context being present when the selected context class differs from the previously determined context class.

62. A mobile device according to claim 60, wherein the processing system is arranged to select, as a response to receiving from the server device data indicating at least one context class determined by one or more classifications carried out in the server device, one of the context classes determined by:

the classification of the vector of feature values,

the classifications of the at least one second vector of feature values, and

the one or more classifications carried out in the server device,

the change of context being present when the selected context class differs from the previously determined context class.

63. A mobile device according to claim 61, wherein the selected context class is the one which has a smallest distance between its ideal feature vector and a corresponding vector of feature values which has been classified, the distance being defined in the sense of a pre-determined vector norm and the ideal feature vector of each context class being such a vector of feature values that best matches with that particular context class.

64. A mobile device according to claim 60, wherein each second vector of feature values is associated with a time limit after which that particular second vector of feature values is no longer valid, and the processing system is arranged to use each second vector of feature values for determining whether the change of context has taken place only if that particular second vector of feature values is still valid.

65. A mobile device according to claim 62, wherein each of the at least one context class determined by the one or more classifications carried out in the server device is associated with a time limit after which that particular context class is no longer valid, and the processing system is arranged to use each of the at least one context class determined by the one or more classifications carried out in the server device for determining, whether the change of context has taken place, only if that particular context class is still valid.

66. A method comprising:

recognizing a change of context related to a mobile device and/or a user of the mobile device,

controlling the mobile device to perform at least one service action related to services provided with the mobile device as a response to a recognized change of context,

gathering usage data indicative of control actions given by the user and directed to the mobile device during different recognized changes of context, and

selecting, prior to performing the at least one service action, the at least one service action from among a pre-determined set of service actions at least partly on the basis of the usage data.

67. A method according to claim 66, wherein the at least one service action is selected using, in addition to the usage data, history data indicative of control actions given by users of other mobile devices during different changes of context related to the other mobile devices and/or the users of the other mobile devices.

68. A method according to claim 66, comprising analyzing, with a data mining algorithm, correlations between different changes of context and different service actions from a data base comprising the usage data.

69. A method according to claim 67, comprising analyzing, with a data mining algorithm, correlations between different changes of context and different service actions from a data base comprising the usage data and the history data.

70. A method according to claim 67, comprising downloading the history data from an external device.

71. A method according to claim 67, comprising receiving the history data via a mobile communication network.

72. A method according to claim 66, comprising obtaining, with one or more sensing entities, sensing data indicative of the context related to the mobile device and/or the user of the mobile device, the change of context being recognized on the basis of the sensing data.

73. A method according to claim 66, comprising receiving, from one or more external device, sensing data indicative of the context related to the mobile device and/or the user of the mobile device, the change of context being recognized on the basis of the sensing data.

74. A method according to claim 72, wherein the recognizing of the change of context comprises:

determining a vector of feature values on the basis of the sensing data, and

classifying the vector of feature values to one of alternative context classes, the change of context being present when the context class determined by the classification of the vector of feature values differs from a previously determined context class.

75. A method according to claim 74, comprising adapting an algorithm used in the classification on the basis of the vector of feature values and feedback information given by the user.

76. A method according to claim 75, wherein the algorithm is adapted so that an ideal feature vector of the context class determined by the classification is moved, depending on the feedback information, closer to or farther away from the vector of feature values, the ideal feature vector of each context class being such a vector of feature values that best matches with that particular context class and the context class determined by the classification being the one whose ideal feature vector is closest to the vector of feature values in the sense of a pre-determined vector norm.

77. A method according to claim 74, wherein the recognizing of the change of context comprises:

checking whether classification of the vector of feature values would fulfill a pre-determined quality criterion,

checking availability of data transfer connections to and from a server device,

as a response to a situation in which the data transfer connections are available and the classification would not fulfill the pre-determined quality criterion, transmitting the vector of feature values to the server device in order to enable the server device to classify the vector of feature values to one of the alternative context classes, and

as a response to receiving a classification result from the server device, using the classification result received from the server device for determining whether the change of context has taken place.

78. A method according to claim 77, comprising calculating the minimum of distances between the vector of feature values and ideal feature vectors of the alternative context classes in order to check whether the classification of the vector of feature values would fulfill the pre-determined quality criterion, the distances being defined in the sense of a pre-determined vector norm and the ideal feature vector of each context class being such a vector of feature values that best matches with that particular context class.

79. A method according to claim 78, comprising transmitting to the server device the ideal feature vector of one or more of the context classes in order to enable the server device to classify the vector of feature values to one of the alternative context classes in a same manner as the classification is carried out in the mobile device.

80. A method according to claim 74, comprising recognizing the change of context on the basis of whether a context class given by the user differs from the previously determined context class.

81. A method according to claim 74, wherein the recognizing of the change of context comprises:

checking availability of data transfer connections to and from the server device,

as a response to a situation in which the data transfer connection is available, transmitting information based on the sensing data to a server device in order to enable the server device to form at least one second vector of feature values, and

as a response to receiving the at least one second vector of feature values from the server device, using each second vector of feature values for determining whether the change of context has taken place.

82. A method according to claim 81, comprising selecting one of the context classes determined by:

the classification of the vector of feature values, and

the classification of each of the at least one second vector of feature values,

the change of context being present when the selected context class differs from the previously determined context class.

83. A method according to claim 81, comprising selecting, as a response to receiving from the server device data indicating at least one context class determined by one or more classifications carried out in the server device, one of the context classes determined by:

the classification of the vector of feature values,

the classifications of the at least one second vector of feature values, and

the one or more classifications carried out in the server device,

the change of context being present when the selected context class differs from the previously determined context class.

84. A method according to claim 82, wherein the selected context class is the one which has a smallest distance between its ideal feature vector and a corresponding vector of feature values which has been classified, the distance being defined in the sense of a pre-determined vector norm and the ideal feature vector of each context class being such a vector of feature values that best matches with that particular context class.

85. A method according to claim 81, wherein each second vector of feature values is associated with a time limit after which that particular second vector of feature values is no longer valid, and each second vector of feature values is used for determining whether the change of context has taken place only if that particular second vector of feature values is still valid.

86. A method according to claim 83, wherein each of the at least one context class determined by the one or more classifications carried out in the server device is associated with a time limit after which that particular context class is no longer valid, and each of the at least one context class determined by the one or more classifications carried out in the server device is used for determining whether the change of context has taken place only if that particular context class is still valid.

87. A non-transitory computer readable medium encoded with a computer program comprising instructions executable by a programmable processor for controlling the programmable processor to:

recognize a change of context related to a mobile device and/or a user of the mobile device,

control the mobile device to perform at least one service action related to services provided with the mobile device as a response to a recognized change of context,

gather usage data indicative of control actions given by the user and directed to the mobile device during different recognized changes of context, and

select, prior to performing the at least one service action, the at least one service action from among a pre-determined set of service actions at least partly on the basis of the usage data.