公开(公告)号：US20230140723A1

公开(公告)日：2023-05-04

申请号：US18089330

申请日：2022-12-27

申请人： Optum Inc.

发明人： Vinod Anand BALASUBRAMANIAN ,  Rama KULASEKARAN ,  Venkatesan SUBRAMANIAN

IPC分类号： G06F21/62 ,  G06F21/60 ,  G06N20/00 ,  G06F9/451 ,  G06Q30/02 ,  H04L9/40

摘要： Embodiments of the present disclosure provide methods, apparatus, systems, computing devices, and computing entities for predictive data protection using a data protection policy determination machine learning model. In one embodiment, a method is provided comprising: processing a historical data corpus using the data protection policy determination machine learning model to generate a dynamic data protection policy update describing inferred data protection instructions; determining an attestation subset of the inferred data protection instructions by comparing the instructions and prior data protection instructions described by an existing data protection policy; for each inferred data protection instruction in the attestation subset, determining a per-instruction attestation determination based on end-user feedback; generating an updated data protection policy by updating the existing policy in accordance with each inferred instruction in the attestation subset whose per-instruction attestation determination describes an affirmative attestation determination; and performing the predictive data protection using the updated data protection policy.

公开(公告)号：US20220129583A1

公开(公告)日：2022-04-28

申请号：US17077571

申请日：2020-10-22

申请人： Optum, Inc.

发明人： Vinod Anand BALASUBRAMANIAN ,  Rama KULASEKARAN ,  Venkatesan SUBRAMANIAN

IPC分类号： G06F21/62 ,  G06F21/60 ,  H04L29/06 ,  G06F9/451 ,  G06Q30/02 ,  G06N20/00

摘要： Embodiments of the present disclosure provide methods, apparatus, systems, computing devices, and computing entities for predictive data protection using a data protection policy determination machine learning model. In one embodiment, a method is provided comprising: processing a historical data corpus using the data protection policy determination machine learning model to generate a dynamic data protection policy update describing inferred data protection instructions; determining an attestation subset of the inferred data protection instructions by comparing the instructions and prior data protection instructions described by an existing data protection policy; for each inferred data protection instruction in the attestation subset, determining a per-instruction attestation determination based on end-user feedback; generating an updated data protection policy by updating the existing policy in accordance with each inferred instruction in the attestation subset whose per-instruction attestation determination describes an affirmative attestation determination; and performing the predictive data protection using the updated data protection policy.