US 12,136,079 B2
Real-time provisioning of targeted recommendations based on decomposed structured messaging data
Christopher Mark Jones, Villanova, PA (US); Barry Wayne Baird, Jr., Kennett Square, PA (US); Claude Bernell Lawrence, Jr., Philadelphia, PA (US); and Jonathan Joseph Prendergast, West Chester, PA (US)
Assigned to The Toronto-Dominion Bank, Toronto (CA)
Filed by The Toronto-Dominion Bank, Toronto (CA)
Filed on Dec. 13, 2021, as Appl. No. 17/549,701.
Claims priority of provisional application 63/126,739, filed on Dec. 17, 2020.
Prior Publication US 2022/0198427 A1, Jun. 23, 2022
Int. Cl. G06Q 30/00 (2023.01); G06Q 20/26 (2012.01)
CPC G06Q 20/26 (2013.01) 22 Claims
OG exemplary drawing
 
16. A computer-implemented method comprising:
receiving, across a first communications channel, and using at least one processor, a message comprising elements of message data disposed within corresponding message fields, the message data characterizing a real-time exchange of data requested by a first counterparty from a second counterparty, the real-time data exchange being associated with a first product provisioned to the second counterparty by the first counterparty, and the first communications channel being established between the apparatus and a computing system associated with the first counterparty;
based on mapping data associated with the message fields, performing operations, using the at least one processor, that obtain at least a subset of the elements of message data from corresponding ones of the message fields, the subset of the elements of message data comprising a first identifier of the first counterparty and a second identifier of the second counterparty;
based on the first and second identifiers, obtaining, using the at least one processor, first data characterizing prior exchanges of data associated with the second counterparty and second data identifying a target parameter that characterizes at least one of the second counterparty or a relationship between the second counterparty and the first counterparty;
based on an application of a trained machine-learning or artificial-intelligence process to an input dataset that includes at least the subset of the elements of message data, a portion of the first data, and a portion of the second data, determining, using the at least one processor, a value of the target parameter that characterizes at least one of the second counterparty or the relationship between the second counterparty and the first counterparty, and obtaining, using the at least one processor, digital content associated with a second product or a service based on the target parameter value; and
transmitting, across a second communications channel, and using the at least one processor, notification data to a device operable by the second counterparty, the notification data comprising the digital content and additional digital content associated with the real-time data exchange, the second communications channel being established between the apparatus and an application program executed by the device, and the notification data causing the executed application program to present a portion of each of the digital content and the additional digital content within a digital interface.