Automated security subsystem activation

1. An apparatus comprising:

a security subsystem;

a processor;

a memory that stores code executable by the processor to:

monitor occupant activity from a plurality of electronic devices;

identify the occupant of the occupant activity;

determine the occupant activity satisfies a quiescence model;

activate the security subsystem in response to satisfying the quiescence model;

determine the occupant activity satisfies an exit model in response to current dress of the occupant conforming to dress data and a location of the occupant being within a proximity threshold of an exit;

determine the occupant has an exit permission for the occupant for a current time, wherein the exit permission specifies a time that the occupant is permitted to exit without activating the security subsystem and the exit permission is based on whether the occupant is an adult or a teenager; and

deactivate the security subsystem in response to both satisfying the exit model and the occupant having exit permission.

2. The apparatus of claim 1 , wherein the exit model is a neural network recursively trained from a plurality of occupant activity data and exit data.

3. The apparatus of claim 1 , wherein the quiescence model is a neural network recursively trained from a plurality of occupant activity data and a last exit.

4. The apparatus of claim 1 , wherein the occupant activity satisfies the quiescence model in response to one or more of a determination that the occupant is asleep, a determination that the occupant is not dressed to exit, a determination that a quiescence motion pattern is satisfied, and a determination that lighting satisfies a quiescence level.

5. The apparatus of claim 1 , wherein the code is further executable by the processor to:

determine a location of each occupant;

forecast an away time for each occupant from an away time model based on the location; and

activate the security subsystem in response to the away time for each occupant exceeding an away time threshold.

6. A method comprising:

monitoring, by use of a processor, occupant activity from a plurality of electronic devices;

identifying the occupant of the occupant activity;

determining the occupant activity satisfies a quiescence model;

activating a security subsystem in response to satisfying the quiescence model;

determining the occupant activity satisfies an exit model in response to current dress of the occupant conforming to dress data and a location of the occupant being within a proximity threshold of an exit;

determining the occupant has an exit permission for the occupant for a current time, wherein the exit permission specifies a time that the occupant is permitted to exit without activating the security subsystem and the exit permission is based on whether the occupant is an adult or a teenager; and

deactivating the security subsystem in response to both satisfying the exit model and the occupant having exit permission.

7. The method of claim 6 , wherein the exit model is a neural network recursively trained from a plurality of occupant activity data and exit data.

8. The method of claim 6 , wherein the quiescence model is a neural network recursively trained from a plurality of occupant activity data and a last exit.

9. The method of claim 6 , wherein the occupant activity satisfies the quiescence model in response to one or more of a determination that the occupant is asleep, a determination that the occupant is not dressed to exit, a determination that a quiescence motion pattern is satisfied, and a determination that lighting satisfies a quiescence level.

10. The method of claim 6 , the method further comprising:

determining a location of each occupant;

forecasting an away time for each occupant from an away time model based on the location; and

activating the security subsystem in response to the away time for each occupant exceeding an away time threshold.

11. A program product comprising a non-transitory computer readable storage medium that stores code executable by a processor, the executable code comprising code to:

monitor occupant activity from a plurality of electronic devices;

identify the occupant of the occupant activity;

determine the occupant activity satisfies a quiescence model;

activate the security subsystem in response to satisfying the quiescence model;

determine the occupant activity satisfies an exit model in response to current dress of the occupant conforming to dress data and a location of the occupant being within a proximity threshold of an exit;

determine the occupant has an exit permission for the occupant for a current time, wherein the exit permission specifies a time that the occupant is permitted to exit without activating the security subsystem and the exit permission is based on whether the occupant is an adult or a teenager; and

deactivate the security subsystem in response to both satisfying the exit model and the occupant having exit permission.

12. The program product of claim 11 , wherein the exit model is a neural network recursively trained from a plurality of occupant activity data and exit data.

13. The program product of claim 11 , wherein the quiescence model is a neural network recursively trained from a plurality of occupant activity data and a last exit.

14. The program product of claim 11 , wherein the occupant activity satisfies the quiescence model in response to one or more of a determination that the occupant is asleep, a determination that the occupant is not dressed to exit, a determination that a quiescence motion pattern is satisfied, and a determination that lighting satisfies a quiescence level.