Systems and methods for evaluating motion event types using machine learning

Abstract

A computerized method for identifying a type of a motion event of a network device is disclosed that includes operations of establishing a position baseline of the network device, wherein establishing the position baseline of the network device includes recording a value for each of an X-, Y- and Z-axis detected by a gyroscope of the network device and detecting the motion event of the network device, which includes detecting a position change by the gyroscope or an acceleration by an accelerometer of the network device. Additionally, the method includes operations of evaluating the detected position change or acceleration of the network device by a machine learning model and identifying a type of motion event based on results of the evaluating by the machine learning model. The method may further include the operation of implementing a set of policies on the network device to restrict functionality thereof.

CPC Classifications