US 12,271,831 B2
Routing a data processing transaction based on machine learning
Pratap Dande, Saint Johns, FL (US); Vijaya L Vemireddy, Plano, TX (US); Brandon Sloane, Indian Land, SC (US); Jinna Z Kim, Charlotte, NC (US); Kapil Chhibber, Davis, CA (US); and Robert Nyeland Huggins, Charlotte, NC (US)
Assigned to Bank of America Corporation, Charlotte, NC (US)
Filed by Bank of America Corporation, Charlotte, NC (US)
Filed on Jun. 4, 2021, as Appl. No. 17/339,686.
Prior Publication US 2022/0391720 A1, Dec. 8, 2022
Int. Cl. G06Q 20/10 (2012.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01); G06Q 20/02 (2012.01)
CPC G06N 5/04 (2013.01) [G06N 20/00 (2019.01); G06Q 20/023 (2013.01); G06Q 20/102 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A routing system, the routing system comprising:
a memory operable to store historical data associated with a plurality of previous data processing transactions; and
a processor operably coupled to a network interface and the memory, and configured to:
obtain the historical data associated with the plurality of previous data processing transactions transacted using a first routing between a first computing system and a second computing system, wherein:
the first computing system is configured to transmit one or more requests to perform one or more data processing transactions to the second computing system; and
the second computing system is configured to receive the one or more requests and perform the one or more data processing transactions;
execute a machine learning algorithm to:
determine one or more entities associated with each previous data processing transaction of the plurality of previous data processing transactions;
assign each of the plurality of previous data processing transactions to a respective class based at least in part upon at least one associated entity;
determine, based on the one or more entities and a plurality of class assignments, a plurality of recurring data processing transactions within the plurality of previous data processing transactions comprising a first entity and a first class, the plurality of recurring data processing transactions being transactions performed by the second computing system that were transacted using the first routing; and
in response to determining the plurality of recurring data processing transactions, generate a second routing between the first computing system and the second computing system configured to route one or more upcoming data processing transactions comprising the first entity and the first class, the second routing being different from the first routing;
determine an upcoming data processing transaction comprising the first entity and the first class; and
use the second routing to transact the upcoming data processing transaction.