US 12,346,966 B2
System and method for transaction reconciliation and anomaly detection using tokenized decentralized mesh
Sakshi Bakshi, New Delhi (IN); Sharat Chandran, Hyderabad (IN); Arif Ahmad Khan, New Delhi (IN); Durga Prasad Kutthumolu, Telangana (IN); Kapil Juneja, Rajasthan (IN); and Arvind Bansal, Haryana (IN)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on May 8, 2023, as Appl. No. 18/144,338.
Prior Publication US 2024/0378668 A1, Nov. 14, 2024
Int. Cl. G06Q 40/04 (2012.01)
CPC G06Q 40/04 (2013.01) [G06Q 2220/00 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computing system comprising:
a real-time trade processing and centralized reconciliation engine comprising:
at least a first processor;
a first communication interface communicatively coupled to the at least first processor; and
first memory storing computer-readable instructions that, when executed by the at least first processor, causes the real-time trade processing and centralized reconciliation engine to:
process a first trade by retrieving first trade metadata associated with the first trade from a first trade system associated with the first trade;
determine whether there are any anomalies in the first trade metadata by performing a centralized reconciliation on the first trade metadata using historical trade data of the first trade system associated with the first trade and an anomaly history of the first trade system associated with the first trade; and
responsive to determining that there are no anomalies in the first trade metadata, send the first trade metadata to an anomaly detection and reconciliation mesh analysis engine; and
the anomaly detection and reconciliation mesh analysis engine, wherein the anomaly detection and reconciliation mesh analysis engine comprises:
at least a second processor;
a second communication interface communicatively coupled to the at least first processor; and
second memory storing computer-readable instructions that, when executed by the at least second processor, causes the anomaly detection and reconciliation mesh analysis engine to:
receive the first trade metadata from the real-time trade processing and centralized reconciliation engine;
tokenize the first trade metadata to generate first tokenized trade metadata;
generate, from at least the first tokenized trade metadata, first tokenized trade digital DNA;
perform hashing on the first tokenized trade digital DNA to generate hashed data;
compare the hashed data to determine whether there are any anomalies within the hashed data; and
responsive to determining that the hashed data comprises at least one anomaly, input the hashed data into a decentralized reconciliation mesh.