Providing ambient information based on learned user context and interaction, and associated systems and devices
Methods, computer readable media, and devices for auto scheduling of ambient information and apps based on learned user context and interaction are described. A method may include recording information corresponding to one or more ambient screens previously displayed to a user, building a probabilistic inference model based at least in part on the recorded information, ranking the one or more ambient screens based at least in part on the probabilistic inference model, selecting a candidate ambient screen from the ranked one or more ambient screens, and displaying the candidate ambient screen during an idle timeslot.
1. A computer-implemented method for selecting and presenting glanceable information by a computerized information system, the method comprising:
recording information corresponding to one or more ambient screens displayed in response to a request of a user, wherein each ambient screen comprises a presentation of information;
building a model for a defined timeslot based at least in part on the recorded information corresponding to the one or more ambient screens displayed in response to the request of the user;
ranking the one or more ambient screens based at least in part on the model;
selecting a candidate ambient screen from the ranked one or more ambient screens; and
displaying the candidate ambient screen during an idle timeslot, wherein, during the idle timeslot, use of the display is not actively being directed by the user.
2. The method of claim 1 , wherein the computerized information system comprises a local device comprising a display screen, and wherein the step of displaying the candidate ambient screen is performed by the local device.
3. The method of claim 2 , wherein the steps of ranking the one or more ambient screens and selecting the candidate ambient screen are performed by a remote computing device that provides the candidate ambient screen to the local device.
4. The method of claim 1 , wherein the information corresponding to the one or more ambient screens comprises one or more elements selected from the group consisting of:
a time of day;
a day of week;
an ambient screen identifier;
a device location;
a user like or dislike;
an environmental condition; and
a user of a local device.
5. The method of claim 1 , wherein recording information corresponding to one or more ambient screens displayed to a user comprises:
receiving an input from the user corresponding to one of the one or more ambient screens displayed, the input received during display of the one or more ambient screens;
recording information corresponding to the one or more ambient screens; and
changing a presentation of information based on the received input.
6. The method of claim 1 , wherein building the model comprises:
determining a size of a schedule time slot;
determining a frequency of switching within a schedule time slot;
determining a boost factor and a reduce factor, wherein the boost factor corresponds to a user preference for an ambient screen and the reduce factor corresponds to a user preference against an ambient screen;
determining a smoothing factor; and
determining one or more weights for smoothing.
7. The method of claim 1 , wherein building the model comprises:
implementing a machine learning algorithm to build the model.
8. The method of claim 1 , wherein the model is a probabilistic inference model.
9. The method of claim 7 , wherein implementing the machine learning algorithm further comprises:
training the machine learning algorithm based on the recorded information corresponding to the one or more displayed ambient screens.
10. The method of claim 1 , wherein selecting the candidate ambient screen from the ranked one or more ambient screens further comprises:
selecting the candidate ambient screen from the ranked one or more ambient screens based on a weighted sampling of the ranked one or more ambient screens.
11. The method of claim 1 , wherein displaying the candidate ambient screen further comprises:
prompting a user for an input selected from the group consisting of:
an option to skip the candidate ambient screen; and
an option to select an alternate candidate screen.
12. The method of claim 11 , wherein displaying the candidate ambient screen further comprises:
recording information corresponding to the candidate ambient screen; and
in response to receipt of a user input, recording the user input as additional information corresponding to the candidate ambient screen.
13. A system for selecting and presenting glanceable information, the system comprising:
a memory configured to store processor instructions; and
a processor in communication with the memory, the processor configured to execute the processor instructions to perform:
recording information corresponding to one or more ambient screens displayed in response to a request of a user, wherein each ambient screen comprises a presentation of information;
building a model for a defined timeslot based at least in part on the recorded information corresponding to the one or more ambient screens displayed in response to the request of the user;
ranking the one or more ambient screens based at least in part on the model;
selecting a candidate ambient screen from the ranked one or more ambient screens; and
causing the candidate ambient screen to be displayed during an idle timeslot, wherein, during the idle timeslot, use of a display is not actively being directed by the user.
14. The system of claim 13 , wherein recording information corresponding to one or more ambient screens displayed to a user comprises:
receiving an input from the user corresponding to one of the one or more ambient screens displayed, the input received during display of the one ambient screen;
recording information corresponding to the one ambient screen; and
changing a presentation of information based on the received input.
15. The system of claim 13 , wherein the processor is configured to execute the processor instructions to perform building the model by implementing a machine learning algorithm to build the model.
16. The system of claim 15 , wherein the model is a probabilistic inference model.
17. A non-transitory computer-readable medium storing computer code for controlling a processor to cause the processor to perform a method for selecting and presenting glanceable information, the computer code including computer instructions to cause the processor to:
record information corresponding to one or more ambient screens displayed in response to a request of a user, wherein each ambient screen comprises a presentation of information;
build a model for a defined timeslot based at least in part on the recorded information corresponding to the one or more ambient screens displayed in response to the request of the user;
rank the one or more ambient screens based at least in part on the model;
select a candidate ambient screen from the ranked one or more ambient screens; and
cause the candidate ambient screen to be displayed during an idle timeslot, wherein, during the idle timeslot, use of a display is not actively being directed by the user.
18. The non-transitory computer-readable medium of claim 17 , wherein the computer code further includes computer instructions to cause the processor to record information corresponding to one or more ambient screens displayed to a user by:
receiving an input from the user corresponding to one of the one or more ambient screens displayed, the input received during display of the one ambient screen;
recording information corresponding to the one ambient screen; and
changing a presentation of information based on the received input.
19. The non-transitory computer-readable medium of claim 17 , wherein the computer code further includes computer instructions to cause the processor to build the model by implementing a machine learning algorithm to build the model.
20. The non-transitory computer-readable medium of claim 19 , wherein the model is a probabilistic inference model.