CPC G06F 8/65 (2013.01) [G06F 8/61 (2013.01); G06F 9/4406 (2013.01); G06F 9/4411 (2013.01); G06F 9/4881 (2013.01); G06F 9/542 (2013.01); G06F 11/3006 (2013.01); G06F 18/2148 (2023.01); G06F 18/24 (2023.01); G06N 20/00 (2019.01)] | 20 Claims |
1. A method, comprising:
registering, by a computer device, devices to a local smart network within a user environment;
collecting, by the computer device, device data from the devices through the local smart network;
compiling, by the computer device, training data from the collected device data;
training, by the computer device, a machine learning model using the training data;
determining, by the computer device, whether a current activity state of each of the devices in the user environment is an active state, wherein a device is currently in use, or an inactive state, wherein a device is not currently in use;
classifying, by the computer device, physical regions of the user environment as an active region or an inactive region based on a current location of a user within the user environment, the current activity states of the devices, and relative locations of the devices with respect to one another in the user environment using the machine learning model;
predicting, by the computer device and using the machine learning model, a time when a target device of the devices will be in an inactive system state based on the classifying the physical regions of the user environment;
checking, by the computer device, whether an update will cause a loss of operation for the target device; and
automatically scheduling, by the computer device, an application of the update for the predicted time when the target device will be in the inactive system state.
|