Context-based interests in computing environments and systems

• US 8,554,767 B2
• Filed: 12/23/2008
• Issued: 10/08/2013
• Est. Priority Date: 12/23/2008
• Status: Expired due to Fees

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Song, Yu, Jeong, Sangoh, Cheng, Doreen, Kalasapur, Swaroop S.
• Primary Examiner(s): Vy, Hung T
• Application Number:
  US12/343,395
• Publication Number:
  US 20100161544A1
• Time in Patent Office: 1,750 Days
• Field of Search: 706/52, 706/45, 706/12, 707732-734, 707/705, 707/722, 707/735, 705/10, 705/7.33, 455/414.1, 455/414.2
• US Class Current: 707/732
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Techniques for determining an interest in an object of interest in a given situation are disclosed. Multiple situation-based interest rating components can be provided for various situations. Each situation-based interest rating component can include an interest value indicative of interest in an object of interest in one of the situations. An input situation can then be obtained. One of the situation-based interest rating components can be identified matching an input situation. The relevance of one or more of the other situation-based interest rating components to the identified matching component can then be determined. This can, for example, be done by computing one or more distances between only the respective situation-based portions of the matching situation-based interest rating component and one or more of the other components, or based on the interest value-based portion of each component, or both.

Claims

1. A computer-implemented method of determining a degree of interest that a user has in a first activity when the user is using a mobile device in a particular context, wherein said method comprises:

obtaining information indicating changes in a context of the user and the device, the information indicating that the user and the device are in a first plurality of different contexts that are selected from the group consisting of

     1) a plurality of different locations that the device and the user are in;

     2) a plurality of different times when the user uses the device;

     3) a plurality of different environments that are sensed by the device;

     4) a plurality of different temperatures that are sensed by the device;

     5) a plurality of different electrical components that are detected by the device;

     6) a plurality of different types of motion that the device senses;

     7) a plurality of different distances between the device and an external object; and

     8) a plurality of different biological conditions of the user that are sensed by the device;

obtaining user data that indicates multiple usages of a device by a user over time, the user data further indicating that each usage of the device involves a particular degree of involvement by the user in the first activity, which further indicates that the user has one of a first plurality of different interest levels in the first activity, the user data further indicating that the user and the device, at the time of each usage, are in one of the first plurality of different contexts;

based on the user data, determining a plurality of context-based interest rating components, each interest rating component associated with a particular context of the first plurality of contexts and including an interest value that estimates a degree of interest of the user in the first activity when the user is in said associated context, each interest rating component being calculated based on the usages of the device that took place in said associated context;

obtaining an input context that indicates that the user and the device are in one of the first plurality of contexts;

identifying, based on said input context, one of said interest rating components as a matching interest rating component for said input context;

determining the relevance of at least a portion of one or more other interest rating components to said matching interest rating component; and

determining, based on said one or more relevancies, a projected interest value indicative of a projected interest of the user in said first activity when the user is in said input context.

2. The method of claim 1 wherein the first plurality of different contexts are the plurality of different locations that the device and user are in.

3. The method of claim 2 wherein:

the user data is organized into a plurality of data associations, each data association providing information about a particular usage of the device by the user, each data association including an associated one of the first plurality of interest levels and an associated one of the plurality of locations, each data association thereby indicating that the user, when using the device at a particular time in the associated one of the plurality of locations, had a degree of interest in the first activity that is represented by the associated one of the interest levels;

each interest rating component is associated with a particular location of the plurality of locations, the interest level of the interesting rating being based on at least some of the data associations of the user data that involve the same location as the particular location associated with the interesting rating component; and

the interest levels of the interest rating components are based on different subsets of the data associations of the user data.

4. The method of claim 3, wherein a first one of the contexts associated with the interest rating components is encoded using 1-in-N encoding, N being an integer equal to the number of possible contexts in the first plurality of contexts, such that the first one of the contexts is encoded as a vector with at least N values.

5. The method of claim 1, wherein said determining of relevance comprises one or more of the following:

determining the relevance of contexts associated with said one or more other interest rating components to the context associated with said matching interest rating component;

determining the relevance of the interest values of said one or more other interest rating components to the interest value of said matching interest rating component; and

determining the relevance of said one or more other interest rating components as a whole to said matching interest rating component considered as a whole.

6. The method of claim 5, wherein each of said interest rating components is represented in a vector form.

7. The method of claim 6, wherein determining the relevance of said one or more other interest rating components as a whole comprises considering entire said vector forms.

8. The method of claim 5, wherein:

at least one of said interest rating components is an unknown interest rating component that includes an unknown interest value; and

said determining of the relevance of interest values of said one or more other interest rating components comprises;

not considering said unknown interest value.

9. The method of claim 1, wherein at least one of said interest rating components is an average-based interest rating component that includes an average interest value.

10. The method of claim 9, wherein said method further comprises:

determining said average interest value at least by averaging a plurality of known interest levels associated with a repeated occurrence of the context associated with said average-based interest rating component in the user data.

11. The method of claim 1, wherein said method further comprises:

assigning an initial value to an unknown interest value, the unknown interest value having a range of possible interest values; and

updating said initial value based on one or more of the following;

a) the projected interest value; and

b) an approximate midpoint of the range of possible interest values.

12. The method of claim 1, wherein:

the user data includes a set of data associations, each data association storing data on each usage of the device by the user, the user data thereby including a plurality of data associations, each data association associated with a particular context of the first plurality of contexts and a particular interest level of the plurality of interest levels, said associated context and said associated interest level indicating that the user was in the associated context when the user had the associated interest level in the first activity, the associated context being represented by at least one data association context variable, each data association context variable having a plurality of possible data association context values.

13. The method of claim 12, wherein said obtaining of said plurality of interest rating components includes averaging the data association interest levels associated with a subset of the set of data associations, the subset including two or more data associations that have the same associated context.

14. The method of claim 1, wherein:

the context associated with a first one of the plurality of interest rating components is encoded as at least a portion of a first multi-dimensional vector;

the input context is encoded as at least a portion of a second multi-dimensional vector;

the determining of the projected interest level uses a weighted sum, the weighted sum being based at least partly on the calculating of a distance between the first multi-dimensional vector and the second multi-dimensional vector.

15. The method of claim 14, wherein the calculating of the distance includes at least one of a group consisting of:

a) calculating a cosine distance; and

b) calculating a Euclidean distance.

16. The method of claim 1, wherein:

each interest rating component is associated with a situation, the interest rating component indicating how much interest the user has in the first activity when the user is in the associated situation; and

the situation associated with each interest rating component is different from the situations associated with any of the others of the plurality of interest rating components, each situation involving at least one of the first plurality of different contexts and one of a second plurality of different contexts.

17. The method of claim 16, wherein the first plurality of different contexts is the plurality of different locations that the device and the user are in and the second plurality of different contexts are a plurality of different times that the user uses the device and wherein the situation of each interest rating component indicates a combination of a particular time and a particular location at which the user uses the device, the combination of said time and said location being unique relative to the combinations associated with the other components in the plurality of interest rating components.

18. The method of claim 1, wherein each one of the plurality of interest rating components includes the interest value that estimates a degree of interest of the user in the first activity and another interest value that estimates a degree of interest of the user in a second, different activity.

19. The method of claim 1 wherein:

a first one of the plurality of situation-based interest rating components is a first unknown interest rating component whose said associated interest value is unknown;

the method further comprises determining a closeness of a match between at least a portion of a first set of one or more of the other interest rating components to at least a portion of said first unknown interest rating component; and

determining, based on the determining of the closeness of the match, a second projected interest value indicative of a projected value for said unknown interest rating.

20. The method of claim 19, further comprising:

substituting said unknown interest value associated with the first unknown interest rating component with the second projected interest value to form an updated interest rating component wherein the one or more other interest rating components used in the determining of the relevance includes the updated interest rating component.

21. A computing system comprising:

at least one processor;

a memory including computer readable instructions that, when executed by the at least one processor, cause the computing system to;

obtain information indicating changes in a context of the user and the device, the information indicating that the user and the device are in a first plurality of different contexts that are selected from the group consisting of

     1) a plurality of different locations that the device and the user are in;

     2) a plurality of different times when the user uses the device;

     3) a plurality of different environments that are sensed by the device;

     4) a plurality of different temperatures that are sensed by the device;

     5) a plurality of different electrical components that are detected by the device;

     6) a plurality of different types of motion that the device senses;

     7) a plurality of different distances between the device and an external object; and

     8) a plurality of different biological conditions of the user that are sensed by the device;

obtain user data that indicates multiple usages of a device by a user over time, the user data further indicating that each usage of the device involves a particular degree of involvement by the user in the first activity, which further indicates that the user has one of a first plurality of different interest levels in the first activity, the user data further indicating that the user and the device, at the time of each usage, are in one of the first plurality of different contexts;

based on the user data, determine a plurality of context-based interest rating components, each interest rating component associated with a particular context of the first plurality of contexts and including an interest value that estimates a degree of interest of the user in the first activity when the user is in said associated context, each interest rating component being calculated based on the usages of the device that took place in said associated context;

obtain an input context that indicates that the user and the device are in one of the first plurality of contexts;

identify based on said input context, one of said interest rating components as a matching interest rating component for said input context;

determine the relevance of at least a portion of one or more other interest rating components to said matching interest rating component; and

determine, based on said one or more relevancies, a projected interest value indicative of a projected interest of the user in said first activity when the user is in said input context.

22. The computing system of claim 21, wherein said computing system includes one or more of the following:

a server configured to perform the determining of the plurality of interest rating components, the identifying of one of said interest rating components, the determining of the relevance, and the determining of the projected interest value; and

a client coupled with the server, the client configured to perform the obtaining of the input context and further configured to transmit the input context to the server.

23. A computer readable medium embodied in a tangible form including executable computer program code operable to predict an interest in an object of interest in a situation, wherein the computer readable medium includes:

executable computer code operable to obtain information indicating changes in a context of the user and the device, the information indicating that the user and the device are in a first plurality of different contexts that are selected from the group consisting of

     1) a plurality of different locations that the device and the user are in;

     2) a plurality of different times when the user uses the device;

     3) a plurality of different environments that are sensed by the device;

     4) a plurality of different temperatures that are sensed by the device;

     5) a plurality of different electrical components that are detected by the device;

     6) a plurality of different types of motion that the device senses;

     7) a plurality of different distances between the device and an external object; and

     8) a plurality of different biological conditions of the user that are sensed by the device;

executable computer code operable to obtain user data that indicates multiple usages of a device by a user over time, the user data further indicating that each usage of the device involves a particular degree of involvement by the user in the first activity, which further indicates that the user has one of a first plurality of different interest levels in the first activity, the user data further indicating that the user and the device, at the time of each usage, are in one of the first plurality of different contexts;

executable computer code operable to determine, based on the user data, a plurality of context-based interest rating components, each interest rating component associated with a particular context of the first plurality of contexts and including an interest value that estimates a degree of interest of the user in the first activity when the user is in said associated context, each interest rating component being calculated based on the usages of the device that took place in said associated context;

executable computer code operable to obtain an input context that indicates that the user and the device are in one of the first plurality of contexts;

executable computer code operable to identify based on said input context, one of said interest rating components as a matching interest rating component for said input context;

executable computer code operable to determine the relevance of at least a portion of one or more other interest rating components to said matching interest rating component; and

executable computer code operable to determine, based on said one or more relevancies, a projected interest value indicative of a projected interest of the user in said first activity when the user is in said input context.

24. A computer-implemented method of determining a degree of interest that a user has in an activity when the user is using mobile device in a particular context wherein said method comprises:

identifying a plurality of different possible contexts in which the user could use the device, wherein the number of possible contexts is predetermined;

identifying a plurality of different possible levels of interest that the user could have in a particular activity, wherein the number of possible interest levels is predetermined;

tracking multiple usages of the mobile device by a user;

during the tracking, for each usage, forming a corresponding usage record;

during the tracking, for each usage, storing information in the corresponding usage record that indicates which context the usage took place in;

during the tracking, for each usage, storing information in the corresponding usage record that relates to how the user interacted with the device;

for each usage, determining that said interaction indicates one of said possible interest levels in said activity, thereby linking each usage of the device and each corresponding usage record with one of the predetermined interest levels and one of the predetermined contexts;

determining a plurality of vectors wherein;

each vector is associated with a different one of the predetermined contexts;

each vector represents an average interest level of the user when the user is in the context associated with the vector;

the average interest level for each vector is an interest level that is calculated based on the interest levels of the relevant usage records in the user data, the relevant usage records for each vector being those usage records that are linked to the context associated with the vector;

obtaining an input context that indicates that the user and the device are in one of the plurality of contexts;

determining a distance between the input context and each of the plurality of vectors that includes a first and second vector wherein the first and second vectors are associated with first and second different contexts and first and second average interest levels, respectively; and

calculating a projected interest level of the user in said activity based on the distance determination wherein if the distance between the first vector and the input context is smaller than the distance between the second vector and the input context, then the first average interest level will be weighed more heavily in the calculation of the projected interest level than the second average interest level,

wherein the plurality of predetermined contexts are selected from the group consisting of;

     1) a plurality of different locations that the device and the user are in;

     2) a plurality of different times when the user uses the device;

     3) a plurality of different environments that are sensed by the device;

     4) a plurality of different temperatures that are sensed by the device;

     5) a plurality of different electrical components that are detected by the device;

     6) a plurality of different types of motion that the device senses;

     7) a plurality of different distances between the device and an external object; and

     8) a plurality of different biological conditions of the user that are sensed by the device.