US 11,768,723 B2
Method and system for predicting failures in interconnected systems based on quantum computing
Venkata Subramanian Jayaraman, Chennai (IN); and Sumithra Sundaresan, Chennai (IN)
Assigned to Wipro Limited, Karnataka (IN)
Filed by Wipro Limited, Bangalore (IN)
Filed on Mar. 27, 2020, as Appl. No. 16/832,062.
Claims priority of application No. 202041006355 (IN), filed on Feb. 13, 2020.
Prior Publication US 2021/0256411 A1, Aug. 19, 2021
Int. Cl. G06F 11/00 (2006.01); G06N 10/00 (2022.01); G06N 20/00 (2019.01); G06F 18/23 (2023.01)
CPC G06F 11/008 (2013.01) [G06F 18/23 (2023.01); G06N 10/00 (2019.01); G06N 20/00 (2019.01)] 15 Claims
OG exemplary drawing
 
1. A method for predicting failures in interconnected systems based on quantum computing, the method comprising:
identifying, by a failure prediction device, a set of unique patterns from input data received from a plurality of input data sources, wherein the plurality of input data sources comprises a plurality of interconnected systems, wherein identifying the set of unique patterns from the input data comprises at least one of processing at least one portion of the input data through a trained Machine Learning (ML) model, when the at least one portion comprises unstructured data, and processing at least one portion of the input data through a data mining algorithm, when the at least one portion comprises structured data;
determining, by the failure prediction device, a correlation between at least two of the plurality of input data sources;
creating, by the failure prediction device, a plurality of sets of clusters corresponding to the plurality of input data sources based on the correlation, wherein each of the plurality of sets of clusters comprises at least two input data sources, and wherein, for each of the plurality of sets of clusters, correlation between corresponding input data sources is above a predefined threshold;
extracting, by the failure prediction device, data associated with each of the set of unique patterns from the input data based on the plurality of sets of clusters;
predicting, by the failure prediction device, based on the extracted data, a failure of at least one interconnected system from the plurality of interconnected systems using the trained ML model;
processing, by the failure prediction device, the extracted data associated with each of the set of unique patterns and information associated with the predicted failure through a quantum computing layer, wherein the quantum computing layer uses a plurality of qubits to process the extracted data and to further identify and respond to an optimized sequence of information; and
generating, by the failure prediction device, through the quantum computing layer, at least one corrective action for the at least one interconnected system.