US 12,081,504 B2
Systems and methods for processing user concentration levels for workflow management
Galen Rafferty, Mahomet, IL (US); Jeremy Goodsitt, Champaign, IL (US); Austin Walters, Savoy, IL (US); and Ernest Kwak, Urbana, IL (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Aug. 31, 2022, as Appl. No. 17/823,954.
Application 17/823,954 is a continuation of application No. 16/824,852, filed on Mar. 20, 2020, granted, now 11,463,391.
Prior Publication US 2023/0060307 A1, Mar. 2, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. H04L 51/212 (2022.01); G06F 16/28 (2019.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01)
CPC H04L 51/212 (2022.05) [G06F 16/285 (2019.01); G06N 3/04 (2013.01); G06N 3/08 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A system for processing user data, the system comprising:
a memory storing instructions; and
at least one processor configured to execute the instructions to perform operations comprising:
receiving, from a client device, first user data associated with a sensor;
filtering the first user data according to a threshold, the threshold comprising a priority level of second user data, the priority level being based on analysis at least one of incoming email content, application messages, notifications from services, Voice over Internet Protocol (VoIP) call messages, or instant messages;
receiving, from a filter model of the client device, feature data corresponding to the first user data, the feature data comprising at least one of workflow information, system messages, or email addresses;
receiving, from the filter model of the client device, a data collection protocol, wherein the data collection protocol includes a command to adjust a hardware setting, modify a data collection parameter, or include a data filter to reject or accept feature data based on a threshold;
training a meta-model to analyze the priority level of the second user data based on the feature data and the data collection protocol, the trained meta-model being configured to determine a statistical relationship or a machine-learned relationship between the data collection protocol and the feature data;
performing, by the trained meta-model, the analysis of the priority level of the second user data;
generating, using the trained meta-model, a meta-model output based on the filter model and the feature data, wherein generating the meta-model output comprises learning from a first data output of a filter model instance and a second data output of a feature model instance;
updating the filter model, based on the meta-model output, to score a concentration level of a user and an importance level of incoming content; and
transmitting the updated filter model to the client device.