US 12,493,884 B2
Systems and methods for predicting illegitimate activity based on user marker levels
Vishi Cline, Frisco, TX (US)
Assigned to Capital One Services, LLC, McLean, VA (US)
Filed by Capital One Services, LLC, McLean, VA (US)
Filed on Jun. 27, 2023, as Appl. No. 18/342,631.
Prior Publication US 2025/0005579 A1, Jan. 2, 2025
Int. Cl. G06Q 20/40 (2012.01)
CPC G06Q 20/4016 (2013.01) 20 Claims
OG exemplary drawing
 
1. A method for predicting illegitimate transaction activity based on detecting user marker levels, the method comprising:
receiving, by a user services system, an identification card comprising a reservoir wherein the reservoir includes a user sample from a user;
determining, by one or more marker sensors of the user services system, a presence of the user sample at the user services system;
collecting, via the one or more marker sensors, sensor device data, the sensor device data including user marker levels from the user sample;
analyzing, via a machine learning model, the user marker levels, wherein the machine learning model has been trained to associate catecholamine levels and epinephrine levels with illegitimate activity to determine a marker range for the user; and
upon determining, by the machine learning model, the user marker levels are above the marker range, automatically rejecting a transaction request of the user and generating and transmitting an illegitimate activity alert to a user device, or
upon determining the user marker levels are below the marker range, automatically generating a request for an additional user sample or an input of additional authentication data.