System and method for determining context
First Claim
Patent Images
1. A method, comprising:
- forming, by at least one processor of a device, a first set of context tags and a second set of context tags, said first set and said second set corresponding to a context of said device, wherein said first set comprises at least one context tag in a first time span of said device and said second set comprises at least one context tag in a second time span of said device,forming, by the at least one processor of the device, a codebook of sparse vectors, said sparse vectors corresponding to said first set of context tags and said second set of context tags, wherein said sparse vectors have a high dimension with respect to a number of non-zero elements in said sparse vectors,using said codebook to form a first sparse context vector from said first set of context tags and a second sparse context vector from said second set of context tags, wherein forming said codebook comprises;
forming random sparse vectors for documents, said documents comprising tags,adding a contribution to a codebook vector of a tag from those said sparse vectors for documents where said tag is present, andrepeating said adding the contribution for a plurality of tags,deriving, by the at least one processor of the device, first derived context data from said first set and second derived context data from said second set,determining, by the at least one processor of the device, a difference between said first derived context data and said second derived context data using said first sparse context vector and said second sparse context vector,determining, by the at least one processor of the device, that a change of context has occurred based on said difference, andcontrolling, by the at least one processor of the device, said device in response to said determined change of context, wherein said second derived context data reflect a change in said context faster than said first derived context data.
4 Assignments
0 Petitions
Accused Products
Abstract
The invention relates to determining a context of a system so that the system can be controlled or the context information be used in the system. The context of a system is described with tags or words that characterize e.g. the state of the system, the surroundings of the system and the state of the user. These tags are determined at multiple time instances. The tags are collected over a first time span and a second time span, and the two (or more) sets of tags are used to determine the context of the system. If the determination using the two sets gives a different result, that is, there is a large distance between the two derived contexts, it is determined that a change in context has occurred. Detecting the change in context can be used to control the system, e.g. by changing the applications presented to the user on the user interface, or by changing the state and/or priority of applications.
-
Citations
16 Claims
-
1. A method, comprising:
-
forming, by at least one processor of a device, a first set of context tags and a second set of context tags, said first set and said second set corresponding to a context of said device, wherein said first set comprises at least one context tag in a first time span of said device and said second set comprises at least one context tag in a second time span of said device, forming, by the at least one processor of the device, a codebook of sparse vectors, said sparse vectors corresponding to said first set of context tags and said second set of context tags, wherein said sparse vectors have a high dimension with respect to a number of non-zero elements in said sparse vectors, using said codebook to form a first sparse context vector from said first set of context tags and a second sparse context vector from said second set of context tags, wherein forming said codebook comprises; forming random sparse vectors for documents, said documents comprising tags, adding a contribution to a codebook vector of a tag from those said sparse vectors for documents where said tag is present, and repeating said adding the contribution for a plurality of tags, deriving, by the at least one processor of the device, first derived context data from said first set and second derived context data from said second set, determining, by the at least one processor of the device, a difference between said first derived context data and said second derived context data using said first sparse context vector and said second sparse context vector, determining, by the at least one processor of the device, that a change of context has occurred based on said difference, and controlling, by the at least one processor of the device, said device in response to said determined change of context, wherein said second derived context data reflect a change in said context faster than said first derived context data. - View Dependent Claims (2, 3, 4, 5, 16)
-
-
6. An apparatus comprising at least one processor, and at least one memory including computer program code for one or more program units, the at least one memory and the computer program code configured to, with the processor, cause the apparatus to perform at least the following:
-
form a first set of context tags and a second set of context tags, said first set and said second set corresponding to a context of said apparatus, wherein said first set comprises at least one context tag in a first time span of said system and said second set comprises at least one context tag in a second time span of said system, form, by the at least one processor of the device, a codebook of sparse vectors, said sparse vectors corresponding to said first set of context tags and said second set of context tags, wherein said sparse vectors have a high dimension with respect to a number of non-zero elements in said sparse vectors, use said codebook to form a first sparse context vector from said first set of context tags and a second sparse context vector from said second set of context tags, wherein forming said codebook comprises; forming random sparse vectors for documents, said documents comprising tags, adding a contribution to a codebook vector of a tag from those said sparse vectors for documents where said tag is present, and repeating said adding the contribution for a plurality of tags, derive first derived context data from said first set and second derived context data from said second set, determine a difference between said first derived context data and said second derived context data using said first sparse context vector and said second sparse context vector, determine that a change of context has occurred based on said difference, and control said apparatus in response to said determined change of context, wherein said second derived context data reflect a change in said context faster than said first derived context data. - View Dependent Claims (7, 8, 9, 10)
-
-
11. A computer program product comprising at least one non-transitory computer-readable medium, said medium including one or more computer-executable instructions that when executed by one or more processors cause a system to carry out at least the following:
-
form a first set of context tags and a second set of context tags, said first set and said second set corresponding to a context of said system, wherein said first set comprises at least one context tag in a first time span of said system and said second set comprises at least one context tag in a second time span of said system, form, by the at least one processor of the device, a codebook of sparse vectors, said sparse vectors corresponding to said first set of context tags and said second set of context tags, wherein said sparse vectors have a high dimension with respect to a number of non-zero elements in said sparse vectors, use said codebook to form a first sparse context vector from said first set of context tags and a second sparse context vector from said second set of context tags, wherein forming said codebook comprises; forming random sparse vectors for documents, said documents comprising tags, adding a contribution to a codebook vector of a tag from those said sparse vectors for documents where said tag is present, and repeating said adding the contribution for a plurality of tags, derive first derived context data from said first set and second derived context data from said second set, determine a difference between said first derived context data and said second derived context data using said first sparse context vector and said second sparse context vector, determine that a change of context has occurred based on said difference, and control said system in response to said determined change of context, wherein said second derived context data reflect a change in said context faster than said first derived context data. - View Dependent Claims (12, 13, 14, 15)
-
Specification
-
- Resources
Litigation Campaign Assessment
Thank you for your request. You will receive a custom alert email when the Litigation Campaign Assessment is available.
×
-
Current AssigneePiece Future Pte., Ltd.
-
Original AssigneeNokia Technologies Oy (Nokia Corporation)
-
InventorsSaarinen, Jukka, Karkkainen, Leo, Terho, Mikko, Fronza, Ilenia, Janes, Andrea, Sillitti, Alberto, Succi, Giancarlo
-
Primary Examiner(s)Pham, Tuan A
-
Application NumberUS14/378,015Publication NumberTime in Patent Office2,435 DaysField of SearchNoneUS Class CurrentCPC Class CodesG06F 16/334 Query execution G06F16/335 ...G06N 20/00 Machine learningG06V 30/10 Character recognitionG06V 30/262 using context analysis, e.g...H04M 1/72454 according to context-relate...
