US 12,222,801 B2
Intelligent issue identifier for auto-detecting dynamic issue changes in a distributed network
Tirupathirao Madiya, Hyderabad (IN); Vishalakshi Nagasai Poosa, Hyderabad (IN); Yellaiah Ponnameni, Hyderabad (IN); Gourav Mohite, Gurugram (IN); and Vinothkumar Babu, Chennai (IN)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on Feb. 2, 2023, as Appl. No. 18/163,791.
Prior Publication US 2024/0264897 A1, Aug. 8, 2024
Int. Cl. G06F 11/00 (2006.01); G06F 11/07 (2006.01); G06N 3/08 (2023.01)
CPC G06F 11/0793 (2013.01) [G06F 11/0709 (2013.01); G06F 11/079 (2013.01); G06N 3/08 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
a memory operable to store:
a plurality of sets of previous data objects associated with corresponding previous application issues and issue patterns associated with corresponding applications, wherein each data object represents an operation status of a corresponding application, wherein each issue pattern represents one or more recurring operation status of the corresponding application, and
a plurality of series of executable operations associated with the corresponding issue patterns configured to solve the previous application issues; and
a processor operably coupled to the memory, the processor configured to:
detect an application issue associated with an application running at a network node at a first timestamp;
receive a first set of data objects associated with the application issue occurring at the first timestamp;
detect the application issue associated with the application running at the network node at a second timestamp;
receive a second set of data objects associated with the application issue occurring at the second timestamp;
determine one or more operation changes between a first set of the data objects and a second set of data objects;
identify, by a machine learning model and based on the one or more operation changes between the first set of the data objects and the second set of data objects a cluster of issue patterns representing the one or more operation changes of the application which occurs between the first timestamp and the second timestamp, wherein each issue pattern in the cluster is associated with a unique strategic solution with a series of executable operations, wherein the machine learning model is trained based on the plurality of sets of previous data objects and corresponding issue patterns associated with the corresponding previous application issues, and wherein the plurality of sets of previous data objects comprises a plurality of operation changes associated with the corresponding application issues;
process, through a neural network, the issue pattern with application information to determine a series of executable operations associated with the application issue, wherein the series of the executable operations is indicative of a solution of the application issue and corresponds to the one or more operation changes of the operation status of the application; and
deploy the series of the executable operations to the application running at the network node to solve the application issue to prevent a failure operation of the application.