Context-based device action prediction
The described implementations relate to automatically performing device actions. One implementation can obtain a contextual value of a contextor. The implementation can decide, using a decision engine, whether to perform an action on a computing device based on the contextual value. In an instance when the decision engine decides that the action is to be performed, the implementation can perform the action on the computing device. The implementation can also update the decision engine using feedback related to the action. As a specific example, the action can be prelaunching an application before a user has requested to execute the application. Prelaunching the application can reduce application latency relative to waiting for the user to request to execute the application before launching the application.
1. A method comprising:
obtaining one or more historical contextual values from one or more contextors; the one or more historical contextual values corresponding to historical instances where a device action was previously initiated by a particular user input provided to a mobile computing device;
training a decision engine to learn a decision rule to automatically perform the device action on the mobile computing device, the decision engine being trained using one or more historical contextual values corresponding to the historical instances where the user provided the particular input to the mobile computing device that initiated the device action;
evaluating one or more current contextual values using the decision rule and determining that the one or more current contextual values trigger the device action; and
in response to the device action being triggered by the one or more current contextual values and in the absence of the particular user input, causing the mobile computing device to automatically perform the device action that was previously initiated by the particular user input.
2. The method according to claim 1 , wherein an individual historical contextual value and an individual current contextual value are both obtained from a physical sensor on the computing device.
3. The method according to claim 1 , wherein an individual historical contextual value reflects activity by a user of the mobile computing device or activity of other users on other mobile computing devices.
4. The method according to claim 3 , the activity comprising the user executing an application on the mobile computing device.
5. The method according to claim 3 , the activity comprising the user communicating with another user using the mobile computing device.
6. The method according to claim 1 , wherein an individual current contextual value reflects a current time or current location of the mobile computing device.
7. The method according to claim 1 , further comprising enabling the action while leaving at least one other action disabled responsive to user interaction with graphical user interface displayed on the mobile computing device.
8. The method according to claim 1 , further comprising enabling an individual contextor while leaving another individual contextor disabled responsive to user interaction with a graphical user interface displayed on the mobile computing device.
9. A mobile computing device comprising:
one or more hardware processors; and
hardware storage storing computer-readable instructions which, when executed by the one or more hardware processors, cause the mobile computing device to:
obtain one or more current contextual values from one or more contextors on the mobile computing device;
provide the one or more current contextual values to a decision engine trained to learn a decision rule that causes the mobile computing device to automatically perform a device action, the decision engine being trained using one or more historical contextual values corresponding to previous instances where explicit user input initiated the device action on the mobile computing device or other mobile computing devices; and
in response to the decision rule being satisfied by the one or more current contextual values and in the absence of the explicit user input, automatically perform the device action on the mobile computing device.
10. The mobile computing device of claim 9 , wherein the computer-readable instructions further cause the mobile computing device to:
train the decision engine using feedback indicating whether the user provided the explicit user input after the device action was automatically performed.
11. The mobile computing device of claim 9 , wherein the computer-readable instructions further cause the mobile computing device to:
obtain the trained decision engine from a server.
12. The mobile computing device of claim 11 , wherein the computer-readable instructions further cause the mobile computing device to:
provide feedback to the server indicating whether the user provided the explicit user input after the device action was automatically performed.
13. The mobile computing device of claim 12 , wherein the computer-readable instructions further cause the mobile computing device to:
provide the one or more current contextual values to the server.
14. The mobile computing device of claim 9 , wherein the device action comprises launching or prelaunching of an application and the explicit user input is a user selection of the application.
15. The mobile computing device of claim 14 , wherein the one or more historical contextual values indicate times and locations when the application was previously launched or prelaunched on the mobile computing device or the other mobile computing devices.