User pattern recognition and prediction system for wearables

• US 10,410,129 B2
• Filed: 12/21/2015
• Issued: 09/10/2019
• Est. Priority Date: 12/21/2015
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Yeung, Fai, Zhou, Fu
• Primary Examiner(s): Chen, Alan
• Application Number:
  US14/976,728
• Publication Number:
  US 20170178011A1
• Time in Patent Office: 1,359 Days
• Field of Search: None
• CPC Class Codes: G06F 1/163 Wearable computers, e.g. on a belt
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 3/01 Input arrangements or combined input and output arrangements for interaction between user and computer G06F3/16 takes precedence
• CPC Class Codes: G06N 5/047 Pattern matching networks; Rete networks
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

One embodiment provides an apparatus. The apparatus includes a companion device. The companion device includes pattern recognition logic to construct a reference graph model based, at least in part, on a plurality of events captured from at least one of the companion device and a wearable device. The reference graph model includes at least one path, each path including one trigger node, at least one event node and a respective edge incident to each event node, a first edge coupling the trigger node and a first event node, a weight associated with each edge corresponding to a likelihood that a second event will follow a first event within a minimum trigger time interval.

Claims

1. An apparatus comprising:

a companion device comprising;

pattern recognition logic to construct a reference graph model based at least in part on a plurality of events captured from at least one of the companion device and a wearable device

the reference graph model comprising at least one path, each path comprising one trigger node, at least one event node and a respective edge incident to each event node, a first edge coupling the trigger node and a first event node, a weight associated with each edge corresponding to a likelihood that a second event will follow a first event within a minimum trigger time interval.

2. The apparatus of claim 1, wherein the companion device further comprises discontinuity detection logic to check for a discontinuity and to replace the reference graph model if a discontinuity is detected or update the reference graph model if a discontinuity is not detected.

3. The apparatus of claim 2, wherein the discontinuity detection logic is to build a test model and to compare the test model to the reference graph model to check for the discontinuity.

4. The apparatus of claim 1, wherein the companion device further comprises prediction logic to predict a next event based at least in part on a current event and on the reference graph model.

5. The apparatus of claim 4, wherein the prediction logic is further to prefetch content associated with the predicted next event.

6. The apparatus of claim 1, wherein each event is selected from the group comprising a user action, a change in sensor data corresponding to a change in state, a device alert and a system notification.

7. An apparatus comprising:

a wearable device comprising;

a reference graph model received from a companion device, the reference graph model constructed based at least in part on a plurality of events captured from at least one of a companion device and the wearable device, the reference graph model comprising at least one path, each path comprising one trigger node, at least one event node and a respective edge incident to each event node, a first edge coupling the trigger node and a first event node, a weight associated with each edge corresponding to a likelihood that a second event will follow a first event within a minimum trigger time interval.

8. The apparatus of claim 7, wherein the wearable device further comprises prediction logic to predict a next event based at least in part on a current event and on the reference graph model.

9. The apparatus of claim 8, wherein the prediction logic is further to prefetch content associated with the predicted next event.

10. The apparatus of claim 7, wherein each event is selected from the group comprising a user action, a change in sensor data corresponding to a change in state, a device alert and a system notification and the state is selected from the group comprising a user activity and a user location.

11. A method comprising:

constructing, by pattern recognition logic, a reference graph model based at least in part on a plurality of events captured from at least one of a companion device and a wearable device,

the reference graph model comprising at least one path, each path comprising one trigger node, at least one event node and a respective edge incident to each event node, a first edge coupling the trigger node and a first event node, a weight associated with each edge corresponding to a likelihood that a second event will follow a first event within a minimum trigger time interval.

12. The method of claim 11, further comprising checking, by discontinuity detection logic, for a discontinuity and replacing, by the discontinuity detection logic, the reference graph model if a discontinuity is detected or updating, by the discontinuity detection logic, the reference graph model if a discontinuity is not detected.

13. The method of claim 12, further comprising building, by the discontinuity detection logic, a test model and comparing, by the discontinuity detection logic, the test model to the reference graph model to check for the discontinuity.

14. The method of claim 11, further comprising predicting, by prediction logic, a next event based at least in part on a current event and the reference graph model.

15. The method of claim 14, further comprising prefetching, by the prediction logic, content associated with the predicted next event.

16. The method of claim 11, wherein each event is selected from the group comprising a user action, a change in sensor data corresponding to a change in state, a device alert and a system notification.

17. A computer readable storage device having stored thereon instructions that when executed by one or more processors result in the following operations comprising:

constructing a reference graph model based, at least in part, on a plurality of events captured from at least one of a companion device and a wearable device,

the reference graph model comprising at least one path, each path comprising one trigger node, at least one event node and a respective edge incident to each event node, a first edge coupling the trigger node and a first event node, a weight associated with each edge corresponding to a likelihood that a second event will follow a first event within a minimum trigger time interval.

18. The device of claim 17, wherein the instructions that when executed by one or more processors results in the following additional operations comprising checking for a discontinuity and replacing the reference graph model if a discontinuity is detected or updating the reference graph model if a discontinuity is not detected.

19. The device of claim 18, wherein the instructions that when executed by one or more processors results in the following additional operations comprising building a test model and comparing the test model to the reference graph model to check for the discontinuity.

20. The device of claim 17, wherein the instructions that when executed by one or more processors results in the following additional operations comprising predicting a next event based at least in part on a current event and on the reference graph model.

21. The device of claim 20, wherein the instructions that when executed by one or more processors results in the following additional operations comprising prefetching content associated with the predicted next event.

22. The device of claim 17, wherein each event is selected from the group comprising a user action, a change in sensor data corresponding to a change in state, a device alert and a system notification.