| CPC H04L 63/08 (2013.01) [H04L 41/145 (2013.01); H04L 41/16 (2013.01); H04L 43/04 (2013.01); H04L 43/062 (2013.01); H04L 63/062 (2013.01); H04L 63/1425 (2013.01); H04L 63/20 (2013.01); H04L 43/16 (2013.01)] | 20 Claims |
|
1. A method, comprising:
receiving, by a computing device, a stream of user data, wherein the stream of user data includes a first set of characteristics associated with the computing device and a second set of characteristics associated with a plurality of user requests received from a user of the computing device;
repeatedly generating, by the computing device using the stream of user data, one or more sets of pre-processed user data, wherein the repeatedly generating is performed according to one or more pre-processing techniques, and wherein the generating includes:
performing one or more feature engineering techniques on characteristics included in the stream of user data; and
repeatedly training, by the computing device using the one or more sets of pre-processed user data, a baseline model to generate a device-trained model, wherein the baseline model is trained at the computing device without providing user data included in the stream of user data to a server computer system;
executing, by the computing device, the device-trained model, wherein executing the device-trained model includes inputting a set of characteristics associated with a user request received from a user of the computing device into the device-trained model; and
transmitting, by the computer device to the server computer system, a score output by the device-trained model for the user request.
|