US 12,299,093 B1
Machine-learning for real-time and secure analysis of digital metrics
Thomas E. Bell, San Francisco, CA (US); Peter Bordow, Fountain Hills, AZ (US); Julio Jiron, San Bruno, CA (US); Akhlaq M. Khan, San Francisco, CA (US); Volkmar Scharf-Katz, San Francisco, CA (US); Jeff J. Stapleton, Arlington, TX (US); Richard Orlando Toohey, San Francisco, CA (US); and Ramesh Yarlagadda, San Francisco, CA (US)
Assigned to Wells Fargo Bank, N.A., San Francisco, CA (US)
Filed by Wells Fargo Bank, N.A., San Francisco, CA (US)
Filed on Aug. 23, 2022, as Appl. No. 17/893,947.
Int. Cl. G06F 21/31 (2013.01); G06F 21/46 (2013.01)
CPC G06F 21/316 (2013.01) [G06F 21/46 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
generating, by a computing system comprising one or more processors, for each digital identity profile in a first set of digital identity profiles, a first metric corresponding to a first time period, and a second metric corresponding to a second time period following the first time period;
identifying, by the computing system, a first subset of the first set of digital identity profiles for which (i) the first metric falls below a threshold, and (ii) the second metric is at least as great as the threshold;
generating, by the computing system, for each digital identity profile in the first subset of digital identity profiles, a training dataset based on a first set of digital identity elements and a second set of digital identity elements;
applying, by the computing system, one or more machine learning models to the training dataset to train an artificial intelligence (AI) agent that is configured to receive, as input, digital identity profiles and provide, as output, transitional elements corresponding to digital identity elements in the digital identity profiles received as input;
generating, by the computing system, a third metric for each digital identity profile in a second set of digital identity profiles;
identifying, by the computing system, a second subset of the second set of digital identity profiles for which the third metric falls below the threshold;
inputting, by the computing system, the second subset of second set of digital identity profiles to the AI agent to generate, for each digital identity profile in the second set of digital identity profiles, a set of transitional elements; and
transmitting, by the computing system, to one or more computing devices identified in the second subset of digital identity profiles, one or more electronic messages corresponding to the set of transitional elements.