US 12,013,748 B2
Intelligent quality accelerator with root mapping overlay
Madhusudhanan Krishnamoorthy, Tamil Nadu (IN); Nagar Parthasarathi Varadarajan, Hyderabad (IN); and Deepika Sehgal, Haryana (IN)
Assigned to BANK OF AMERICA CORPORATION, Charlotte, NC (US)
Filed by BANK OF AMERICA CORPORATION, Charlotte, NC (US)
Filed on Jun. 9, 2021, as Appl. No. 17/342,678.
Prior Publication US 2022/0398523 A1, Dec. 15, 2022
Int. Cl. G06F 11/00 (2006.01); G06F 11/07 (2006.01); G06F 16/22 (2019.01); G06F 16/242 (2019.01); G06F 16/2455 (2019.01); G06N 3/049 (2023.01); G06Q 10/0639 (2023.01)
CPC G06F 11/079 (2013.01) [G06F 16/221 (2019.01); G06F 16/2445 (2019.01); G06F 16/24564 (2019.01); G06N 3/049 (2013.01); G06Q 10/06395 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A system for an intelligent quality accelerator with root mapping, the system comprising:
a memory device with computer-readable program code stored thereon;
a communication device;
a processing device operatively coupled to the memory device and the communication device, wherein the processing device is configured to execute the computer-readable program code to:
present a rules interface for sample input extraction from one or more product databases;
receive user selected rules for sample input extraction and extract sample inputs from product databases in accordance with user selected rules and a module comprising a business value metric (BVM), a language model, and a private data governance model to generate sample rules, wherein the BVM provides a weighted column to each database column that weights columns within the product database with respect to the user selected rules;
convert the sample rules into sequel statements and apply the sequel statements against the product databases to extract the sample inputs;
convert the sample inputs into graphical format and overlay the sample input against a current resource exchange;
identify a node of divergence between the graphical format of the sample inputs and the current resource exchange;
translate the node of divergence to a vector for root cause identification, wherein translating the node of divergence to the vector for root cause identification further comprises transmitting graphical data to a decoder to decode via a long short term memory model (LSTM), wherein the LSTM model identifies a branch where nodes are deviating and identifies a root cause of the divergence; and
present the root cause identification to the user along with a recommendation based on historic tested results.