US 12,412,412 B1
Unstructured data identification and workflow execution using machine-learning techniques
Jason W. Black, Columbus, OH (US); Timothy Gorman, Columbus, OH (US); and Carrie A. Kubasta, Columbus, OH (US)
Assigned to The Huntington National Bank, Columbus, OH (US)
Filed by THE HUNTINGTON NATIONAL BANK, Columbus, OH (US)
Filed on Jan. 13, 2025, as Appl. No. 19/018,479.
Int. Cl. G06V 30/19 (2022.01); G06N 20/00 (2019.01); G06V 10/70 (2022.01)
CPC G06V 30/19147 (2022.01) [G06N 20/00 (2019.01); G06V 10/70 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
obtaining, by a computing device, a machine-learning model that has been trained to identify a set of textual data instances within an input image and data types corresponding to the set of textual data instances, the machine-learning model being previously trained utilizing a machine-learning algorithm and a training data set comprising example images of a plurality of unstructured formats, each example identifying corresponding textual data instances within an image of an unstructured format of the plurality of unstructured formats and corresponding data types of the corresponding textual data instances;
receiving, by the computing device, a first image having a first unstructured format;
providing, by the computing device, the first image as input to the machine-learning model;
obtaining, by the computing device, first output from the machine-learning model, the first output identifying a first set of textual data instances identified within the first image and a first set of data types corresponding to the first set of textual data instances;
obtaining, by the computing device, data associated with a data provider by identifying a first textual data instance of the first set of textual data instances based at least in part on a first data type identified for the first textual data instance;
determining, by the computing device, whether a second textual data instance of the first set of textual data instances is valid based at least in part on comparing the second textual data instance to the data associated with the data provider;
generating, by the computing device, a message comprising one or more values corresponding to the first set of textual data instances and a transaction authorization for an account associated with the first image; and
based at least in part on determining that the second textual data instance is valid, transmitting, by the computing device, the message comprising the one or more values corresponding to the first set of textual data instances, the message being transmitted as part of executing an automated process.