US 12,444,010 B2
Machine learning modeling for protection against online disclosure of sensitive data
Irgelkha Mejia, Round Rock, TX (US); Ronald Oribio, Austin, TX (US); Robert Burke, Austin, TX (US); and Michele Saad, Austin, TX (US)
Assigned to Adobe Inc., San Jose, CA (US)
Filed by Adobe Inc., San Jose, CA (US)
Filed on Oct. 18, 2023, as Appl. No. 18/489,399.
Application 18/489,399 is a continuation of application No. 17/093,175, filed on Nov. 9, 2020, granted, now 11,830,099.
Prior Publication US 2024/0046399 A1, Feb. 8, 2024
This patent is subject to a terminal disclaimer.
Int. Cl. G06Q 10/10 (2023.01); G06F 16/48 (2019.01); G06F 21/62 (2013.01); G06F 40/40 (2020.01); G06N 3/08 (2023.01); G06Q 10/06 (2023.01); G06Q 10/0635 (2023.01); G06Q 30/02 (2023.01); G06Q 30/06 (2023.01); G06Q 40/08 (2012.01); G06Q 50/00 (2012.01); G06Q 50/26 (2012.01); G06F 3/0482 (2013.01)
CPC G06Q 50/265 (2013.01) [G06F 16/48 (2019.01); G06F 21/6245 (2013.01); G06F 40/40 (2020.01); G06N 3/08 (2013.01); G06Q 10/0635 (2013.01); G06Q 10/10 (2013.01); G06Q 50/01 (2013.01); G06F 3/0482 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
identifying, with a natural language processing subsystem, a plurality of entities associated with private information by at least applying a trained machine learning model to a set of unstructured text data received from a graphical interface;
computing, by a scoring subsystem, a privacy score for the text data by identifying connections between the entities, the connections between the entities contributing to the privacy score according to a cumulative privacy risk, accounting for the risk of exposing certain entities together, the privacy score indicating potential exposure of the private information by the set of unstructured text data; and
updating, by a reporting subsystem in real time, the graphical interface to include an indicator distinguishing a target portion of the set of unstructured text data from other portions of the set of unstructured text data, wherein a modification to the target portion changes the potential exposure of the private information indicated by the privacy score,
wherein the machine learning model includes a neural network, the method further comprising training the neural network by:
retrieving, by a training subsystem, first training data for a first entity type associated with privacy risk from a first database;
retrieving, by the training subsystem, second training data for a second entity type associated with privacy risk from a second database; and
training, by the training subsystem, the neural network to identify the first entity type and the second entity type using the first training data and the second training data.