Systems and methods for determining data criticality based on causal evaluation

Patent metadata

Field	Value
Publication number	US-12505364-B2
Application number	US-202117236044-A
Country	US
Kind code	B2
Filing date	Apr 21, 2021
Priority date	Apr 21, 2021
Publication date	Dec 23, 2025
Grant date	Dec 23, 2025

Abstract

Techniques described herein relate to methods and systems for determining data asset criticality. Such techniques may include making a first determination that a plurality of data asset attributes are part of a causal attribute set; calculating a SHapeley Additive explanation (SHAP) value for each of the plurality of data asset attributes in the causal attribute set; and performing a weighted mean calculation using the SHAP values for each of the plurality of data asset attributes and a corresponding attribute value for each of the plurality of data asset attributes of a data asset to obtain a criticality score for the data asset.

First claim

What is claimed is:

1 . A method for determining data asset criticality, the method comprising:
making a first determination that a plurality of data asset attributes are part of a causal attribute set; calculating a SHapley Additive explanation (SHAP) value for each of the plurality of data asset attributes in the causal attribute set; performing a weighted mean calculation using the SHAP values for each of the plurality of data asset attributes and a corresponding attribute value for each of the plurality of data asset attributes of a data asset to obtain a criticality score for the data asset; including the data asset in a ranking of data assets using the criticality score; and classifying the data asset into a criticality classification based on a criticality threshold; generating a backup recommendation set based on the criticality classification,
wherein the backup recommendation set comprises backup actions comprising at least one selected from a group consisting of a key rotation frequency adjustment, a backup retention adjustment, a backup storage type adjustment, and a backup storage location adjustment; and
adjusting an attribute of the data asset to include a backup action of the backup actions.

2 . The method of claim 1 , wherein making the first determination that the plurality of data asset attributes are part of the causal attribute set comprises:
performing a data category value calculation using a historical attribute set of a plurality of data assets to obtain a plurality of data category values for a plurality of data assets; and performing a causal inference analysis using the historical attribute set, a causal graph, and the plurality of data category values to obtain the causal attribute set.

3 . The method of claim 2 , wherein the data category value calculation comprises a linear regression analysis.

4 . The method of claim 2 , wherein the causal graph is a directed acyclic graph (DAG).

5 . The method of claim 1 , wherein, before the weighted mean calculation, the SHAP values are scaled to values between zero and one.

6 . A non-transitory computer readable medium comprising computer readable program code, which when executed by a computer processor enables the computer processor to perform a method for determining data asset criticality, the method comprising:
making a first determination that a plurality of data asset attributes are part of a causal attribute set; calculating a SHapley Additive explanation (SHAP) value for each of the plurality of data asset attributes in the causal attribute set; and performing a weighted mean calculation using the SHAP values for each of the plurality of data asset attributes and a corresponding attribute value for each of the plurality of data asset attributes of a data asset to obtain a criticality score for the data asset; including the data asset in a ranking of data assets using the criticality score; and classifying the data asset into a criticality classification based on a criticality threshold; generating a backup recommendation set based on the criticality classification,
wherein the backup recommendation set comprises backup actions comprising at least one selected from a group consisting of a key rotation frequency adjustment, a backup retention adjustment, a backup storage type adjustment, and a backup storage location adjustment; and
adjusting an attribute of the data asset to include a backup action of the backup actions.

7 . The non-transitory computer readable medium of claim 6 , wherein making the first determination that the plurality of data asset attributes are part of the causal attribute set comprises:
performing a data category value calculation using a historical attribute set of a plurality of data assets to obtain a plurality of data category values for a plurality of data assets; and performing a causal inference analysis using the historical attribute set, a causal graph, and the plurality of data category values to obtain the causal attribute set.

8 . The non-transitory computer readable medium of claim 7 , wherein the data category value calculation comprises a linear regression analysis.

9 . The non-transitory computer readable medium of claim 7 , wherein the causal graph is a directed acyclic graph (DAG).

10 . The non-transitory computer readable medium of claim 6 , wherein, before the weighted mean calculation, the SHAP values are scaled to values between zero and one.

11 . A system for determining data asset criticality, the system comprising:
a data valuator, comprising a processor, memory, and a storage device, operatively connected to a plurality of data assets, and configured to: make a first determination that a plurality of data asset attributes are part of a causal attribute set; calculate a SHapley Additive explanation (SHAP) value for each of the plurality of data asset attributes in the causal attribute set; and perform a weighted mean calculation using the SHAP values for each of the plurality of data asset attributes and a corresponding attribute value for each of the plurality of data asset attributes of a data asset to obtain a criticality score for the data asset; include the data asset in a ranking of data assets using the criticality score; and classify the data asset into a criticality classification based on a criticality threshold; generate a backup recommendation set based on the criticality classification,
wherein the backup recommendation set comprises backup actions comprising at least one selected from a group consisting of a key rotation frequency adjustment, a backup retention adjustment, a backup storage type adjustment, and a backup storage location adjustment; and
adjust an attribute of the data asset to include a backup action of the backup actions.

12 . The system of claim 11 , wherein, to make the first determination that the plurality of data asset attributes are part of the causal attribute set, the data valuator is further configured to:
perform a data category value calculation using a historical attribute set of a plurality of data assets to obtain a plurality of data category values for a plurality of data assets; and perform a causal inference analysis using the historical attribute set, a causal graph, and the plurality of data category values to obtain the causal attribute set.

13 . The system of claim 12 , wherein the data category value calculation comprises a linear regression analysis, and the causal graph is a directed acyclic graph (DAG).

14 . The system of claim 11 , wherein, before the weighted mean calculation, the SHAP values are scaled to values between zero and one.

Assignees

• Emc Ip Holding Co Llc

Inventors

• Chopra Shelesh
• Vishwakarma Rahul Deo

Classifications

• G06F18/2135

  based on approximation criteria, e.g. principal component analysis · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06F11/1461

  Backup scheduling policy · CPC title

• G06N20/00

  Machine learning · CPC title