US 12,438,868 B2
Pre-access automated login to computing devices, applications and web sites/services based on machine-learning determined user access patterns
Maharaj Mukherjee, Poughkeepsie, NY (US); and George Anthony Albero, Charlotte, NC (US)
Assigned to BANK OF AMERICA CORPORATION, Charlotte, NC (US)
Filed by BANK OF AMERICA CORPORATION, Charlotte, NC (US)
Filed on Jan. 2, 2024, as Appl. No. 18/402,163.
Prior Publication US 2025/0220014 A1, Jul. 3, 2025
Int. Cl. H04L 9/40 (2022.01); G06F 21/31 (2013.01); G06F 21/41 (2013.01); H04L 67/02 (2022.01)
CPC H04L 63/083 (2013.01) [G06F 21/31 (2013.01); G06F 21/41 (2013.01); H04L 63/0815 (2013.01); H04L 67/02 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A system for computing login management, the system comprising:
one or more Machine-Learning (ML) models trained to learn login patterns for a user, wherein the login patterns including at least one of login for computing devices, networks, applications and web sites/services; and
a first computing apparatus including a first memory and one or more first computing processor devices in communication with the first memory, wherein the first memory stores:
a passcode vault that stores a plurality of passcodes for the user, wherein each of the plurality of passcodes is associated with one chosen from the group consisting of (i) a computing device, (ii) a network, (iii) an application and (iv) a web site/service; and
a login management application that is executable by at least one of the one or more first computing processor devices and is configured to:
receive a first user input that defines a first computing apparatus login passcode for the first computing apparatus;
map the first computing apparatus login passcode to one or more first user characteristics previously captured or identified and store said mapping within the first memory;
verify that the first user input matches a predefined apparatus login passcode;
in response to verifying that the first user input matches the predefined apparatus login passcode, capture or identify current user characteristics;
verify that the current user characteristics match the first user characteristics;
in response to verifying that the first user input matches the predefined apparatus login passcode and that the current user characteristics match the first user characteristics, unlock the first computing apparatus and the passcode vault;
in response to unlocking the passcode vault, implement at least one of the one or more ML models, using a first computing apparatus identifier as an input, to identify at least one login pattern from amongst the learned login patterns;
retrieve, from the unlocked passcode vault, first passcodes from amongst the plurality of passcodes, wherein the first passcodes are associated with at least one of computing devices, networks, applications and web sites/services in the at least one login pattern; and
prior to the user initiating access to the at least one of computing devices, networks, applications and web sites/services for which the first passcodes are associated with, access the at least one of computing devices, networks, applications and web sites/services and apply the first passcodes at the at least one of computing devices, networks, applications and web sites/services in a sequence defined by the at least one login pattern to grant the user pre-access authorization to the least one of the computing devices, the applications, the networks, the web sites/services absent further user input after the first user input.