Abstract: Embodiments of the present invention provide techniques, systems, and methods for crowdsourcing device recognition to collect device information and identification data from a limited number of network devices and then leverage the collected information with machine learning techniques to expand the starting set in way that the prediction of device attributes like device type, device brand, family and model can be applied on billions of devices.

Abstract: Embodiments of the present invention provide techniques, systems, and methods for determining and modifying a device timeout of the connection of a device on a network and for using a coefficient of adhesion to determine device state data. Network protocols utilized by the device on the network may be determined based on a determined category of the device. A device timeout may be calculated based on the category and the network protocols. Device data, including state data of the device on the network may be obtained, and analyzed to determine at least one statistic value for the device state data. The device timeout may be modified based on the at least one statistic value. In addition, device data, corresponding to the device may be obtained and analyzed to determine a coefficient of adhesion. State data of the device on the network may be determined based on the coefficient of adhesion.