公开(公告)号：US20250112941A1

公开(公告)日：2025-04-03

申请号：US18376130

申请日：2023-10-03

Applicant: BANK OF AMERICA CORPORATION

Inventor： John Howard Kling ,  Charles Edward Dudley ,  Jason T. Yeung

IPC: H04L9/40

Abstract: Systems, methods, and computer program products are provided herein for data security model modification and anomaly detection. An example method includes receiving a model input associated with one or more data entries of a data security model and accessing the data security model. The data security model includes a plurality of data objects including one or more data entries where each data object defines an associated model level indicative of the hierarchical position of the data object within the data security model and one or more links between the data objects that define data object interdependency parameters. The example method further includes determining one or more data objects of the data security model implicated by the model input and modifying one or more data entries of the one or more implicated data objects of the data security model in response to the model input.

公开(公告)号：US20220131762A1

公开(公告)日：2022-04-28

申请号：US17077615

申请日：2020-10-22

Applicant: BANK OF AMERICA CORPORATION

Inventor： Mark Earl Brubaker ,  Jason T. Yeung

IPC: H04L12/24 ,  G06K9/62 ,  G06N20/00 ,  H04L12/26 ,  H04L29/06

Abstract: Systems, computer program products, and methods are described herein for real-time imitation network generation using artificial intelligence. The present invention is configured to electronically receive, from a computing device of a user, a real dataset; initiate one or more machine learning algorithms on the real dataset; determine, using the one or more machine learning algorithms, one or more data distribution parameters associated with the real dataset; electronically receive, from the computing device of the user, a first shift parameter; skew the one or more data distribution parameters using the first shift parameter to generate one or more skewed data distribution parameters; and generate, using the one or more machine learning algorithms, an imitation dataset using the one or more skewed data distribution parameters.

公开(公告)号：US20250165612A1

公开(公告)日：2025-05-22

申请号：US18516873

申请日：2023-11-21

Applicant: BANK OF AMERICA CORPORATION

Inventor： John Howard Kling ,  Charles Edward Dudley ,  Jason T. Yeung

IPC: G06F21/57 ,  G06F21/64

Abstract: Systems, methods, and computer program products are provided herein for data security model based anomaly determinations. An example method includes receiving a product evaluation request that is associated with a first product dataset including product data entries and accessing a data security model. The data security model includes a plurality of data objects including one or more data entries where each data object defines an associated model level indicative of the hierarchical position of the data object within the data security model and one or more links between the data objects that define data object interdependency parameters. The example method includes determining data objects of the data security model applicable to the first product dataset and determining one or more anomalies associated with the first product dataset based on a comparison between the one or more product data entries and the applicable data objects of the data security model.

公开(公告)号：US11522767B2

公开(公告)日：2022-12-06

申请号：US17077615

申请日：2020-10-22

Applicant: BANK OF AMERICA CORPORATION

Inventor： Mark Earl Brubaker ,  Jason T. Yeung

IPC: H04L29/08 ,  H04L41/14 ,  G06K9/62 ,  H04L9/40 ,  H04L41/16 ,  H04L43/045 ,  H04L43/50 ,  G06N20/00 ,  G06V10/75

Abstract: Systems, computer program products, and methods are described herein for real-time imitation network generation using artificial intelligence. The present invention is configured to electronically receive, from a computing device of a user, a real dataset; initiate one or more machine learning algorithms on the real dataset; determine, using the one or more machine learning algorithms, one or more data distribution parameters associated with the real dataset; electronically receive, from the computing device of the user, a first shift parameter; skew the one or more data distribution parameters using the first shift parameter to generate one or more skewed data distribution parameters; and generate, using the one or more machine learning algorithms, an imitation dataset using the one or more skewed data distribution parameters.