US 12,223,258 B2
Rules/model-based data processing system for intelligent event prediction in an electronic data interchange system
Roger David Moyers, Spring Hill, TN (US)
Assigned to Open Text GXS ULC, Halifax (CA)
Filed by Open Text GXS ULC, Halifax (CA)
Filed on May 1, 2023, as Appl. No. 18/310,409.
Application 18/310,409 is a continuation of application No. 17/548,407, filed on Dec. 10, 2021, granted, now 11,699,025.
Application 17/548,407 is a continuation of application No. 17/171,546, filed on Feb. 9, 2021, granted, now 11,200,370, issued on Dec. 14, 2021.
Application 17/171,546 is a continuation of application No. 16/789,089, filed on Feb. 12, 2020, granted, now 10,922,477, issued on Feb. 16, 2021.
Application 16/789,089 is a continuation of application No. 15/895,693, filed on Feb. 13, 2018, granted, now 10,585,979, issued on Mar. 10, 2020.
Prior Publication US 2023/0267269 A1, Aug. 24, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 40/16 (2020.01); G06F 16/35 (2019.01); G06F 40/123 (2020.01); G06F 40/137 (2020.01); G06N 20/00 (2019.01)
CPC G06F 40/16 (2020.01) [G06F 16/35 (2019.01); G06F 40/123 (2020.01); G06F 40/137 (2020.01); G06N 20/00 (2019.01)] 23 Claims
OG exemplary drawing
 
1. A system for electronic data interchange (EDI) management comprising:
a networked EDI system configured to receive EDI documents over a network, capture EDI document data and deliver the EDI documents over the network;
a memory configured to store data comprising:
EDI document data;
a machine learning model representing element information of a corpus of training EDI documents of a first type and metrics data that classifies the training EDI documents of the first type, the machine learning model trained to classify documents according to a plurality of classes;
a processor that is configured to perform a method comprising:
for an EDI document of the first type:
extracting elements from the EDI document of the first type and creating a document record for the EDI document of the first type, the document record comprising the elements extracted from the EDI document of the first type;
predicting a class from the plurality of classes for the EDI document of the first type by processing the extracted elements using the machine learning model; and
adding a label for the predicted class to the document record for the EDI document of the first type, the label accessible to a client computer via a presentation layer.