US 12,135,638 B2
Graphical neural network for error identification
Sakshi Bakshi, New Delhi (IN); Srinivasa Dhanwada, Hyderabad (IN); Sathya Thamilarasan, Tamilnadu (IN); Siva Paini, Telangana (IN); Shyamala Manoharan, Chennai (IN); Nagalaxmi Sama, Telangana (IN); and Sri Lakshmi Priya Doraiswamy, Charlotte, NC (US)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on Oct. 25, 2022, as Appl. No. 17/972,791.
Prior Publication US 2024/0134777 A1, Apr. 25, 2024
Prior Publication US 2024/0232054 A9, Jul. 11, 2024
Int. Cl. G06F 11/36 (2006.01); G06F 11/34 (2006.01)
CPC G06F 11/3668 (2013.01) [G06F 11/3457 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computing platform comprising:
at least one processor;
a communication interface communicatively coupled to the at least one processor; and
memory storing computer-readable instructions that, when executed by the at least one processor, cause the computing platform to:
receive system parameters of an enterprise system that represent characteristics of the enterprise system and virtual parameters of a simulated enterprise system, in which an application has been upgraded, that represent characteristics of the simulated enterprise system and correspond to the system parameters, wherein:
each virtual parameter that represents a characteristic of the simulated enterprise system corresponds to a system parameter that represents a corresponding characteristic of the enterprise system; and
differences between values of the virtual parameters and values of the corresponding system parameters represent differences between the characteristics of the simulated enterprise system and the corresponding characteristics of the enterprise system;
store, in a graphical database, the system parameters and the virtual parameters;
create system nodes based on the system parameters and virtual nodes based on the virtual parameters, wherein the system nodes are linked together based on relationships between the system parameters, and the virtual nodes are linked together based on relationships between the virtual parameters;
dynamically link the system nodes to the virtual nodes based on the correspondence between the system parameters and the virtual parameters;
detect, based on the dynamic linking between the system nodes and the virtual nodes, at least one error, wherein detecting the at least one error comprises identifying that at least one dynamic link indicates a discrepancy between a system node and a corresponding virtual node;
input the at least one error into an AI engine to output at least one action to correct the at least one error;
modify the graphical database by updating the virtual nodes based on the at least one action to correct the at least one error;
dynamically link the system nodes to the updated virtual nodes;
determine whether the at least one action corrected the at least one error based on the dynamic linking of the system nodes to the updated virtual nodes; and
based on identifying that the at least one action corrected the at least one error, send one or more commands directing the enterprise system to upgrade the application within the enterprise system and execute the at least one action to correct the at least one error on the enterprise system, wherein sending the one or more commands causes the enterprise system to upgrade the application and execute the at least one action to correct the at least one error.