US 11,915,246 B2
Methods and systems for providing a decision making platform
John D. Chisholm, Ballwin, MO (US)
Assigned to MASTERCARD INTERNATIONAL INCORPORATED, Purchase, NY (US)
Filed by MASTERCARD INTERNATIONAL INCORPORATED, Purchase, NY (US)
Filed on Mar. 14, 2022, as Appl. No. 17/694,315.
Application 17/694,315 is a continuation of application No. 16/505,397, filed on Jul. 8, 2019, granted, now 11,276,066.
Application 16/505,397 is a continuation of application No. 14/293,734, filed on Jun. 2, 2014, granted, now 10,346,844, issued on Jul. 9, 2019.
Application 14/293,734 is a continuation in part of application No. 13/364,190, filed on Feb. 1, 2012, granted, now 8,744,941, issued on Jun. 3, 2014.
Application 13/364,190 is a continuation of application No. 12/271,643, filed on Nov. 14, 2008, granted, now 8,126,791, issued on Feb. 28, 2012.
Prior Publication US 2022/0222672 A1, Jul. 14, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 20/40 (2012.01); G06Q 20/04 (2012.01); G06Q 40/02 (2023.01)
CPC G06Q 20/4016 (2013.01) [G06Q 20/04 (2013.01); G06Q 20/40 (2013.01); G06Q 40/02 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for applying a plurality of computer-implemented prediction models to requests received via a plurality of input data streams, said method performed by a computer system comprising at least one processor and a memory device in communication with the at least one processor, the at least one processor in communication with a payment card processing network, said method comprising steps executed by the at least one processor of:
receiving the requests via the plurality of input data streams, wherein the plurality of input data streams includes at least two of a) real-time authorization requests routed individually from the payment card processing network, b) authorization advice messages routed individually from the payment card processing network after completion of an authorization process, and c) batch files of authorization advice messages received via a file transfer protocol;
creating, for each of the received requests, a corresponding data object having a common data structure across the received requests from the plurality of input data streams, wherein each data object contains transaction data extracted from the corresponding received request;
associating each data object with an execution plan specifying a corresponding at least one of the plurality of prediction models;
invoking, for each data object, the corresponding at least one of the prediction models specified by the execution plan, wherein each prediction model operates on the transaction data in the data object and returns output data;
converting, for each of the received requests, the output data returned from the corresponding at least one prediction model for the associated data object into compatible output data in a format appropriate for transmission to one or more intended recipients; and
transmitting the compatible output data for each received request to the one or more intended recipients.