Automated Security Policy Tagging via Content Feature Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Manually creating and managing security policies for documents is time-consuming and inconsistent, requiring manual designation of authorizations and permissions, which can be erroneous and does not scale with increasing document volume.

Innovation Solution

A security policy prediction model is used to automatically identify and suggest relevant security policies for new and existing documents by analyzing content features and historical access data, allowing for automated tagging and modification suggestions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If security policies are manually created by document authors or administrators, then authorized users and permissions can be designated, but the process becomes time-consuming and inconsistent

Engineering Contradiction:

Improveconsistency of security policy applicationVSAvoidtime required to create security policies

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system enables documents to automatically tag themselves with relevant security policies by analyzing their own content features and comparing them against the security policy data store, eliminating the need for manual intervention while maintaining consistent policy application

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The manual mechanical process of creating security policies is replaced with an automated computational system that uses content feature extraction, similarity comparison algorithms, and automatic tagging to generate security policies

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Adaptability or versatility

If manual security policy designation is performed, then specific authorizations and permissions can be assigned, but errors may occur and scalability is limited

Engineering Contradiction:

Improvescalability with document volumeVSAvoidaccuracy of security policy assignment

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system allows documents to autonomously identify and tag themselves with appropriate security policies through automated content analysis and similarity matching, ensuring accurate and consistent policy assignment across growing document volumes without manual intervention

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system incorporates user feedback when authors review and confirm suggested security policies, using this feedback to continuously improve the accuracy of automatic tagging through refined similarity comparisons and learning from correction patterns

Inventive Principle:

Principle #23Feedback

3Ease of operation

If authors must understand document contents to determine authorized users, then appropriate security policies can be created, but the process becomes complex and time-consuming

Engineering Contradiction:

Improveease of security policy creationVSAvoidtime required to analyze document and create policy

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The document automatically performs the analysis of its own content and identification of appropriate security policies through automated feature extraction and comparison algorithms, freeing the author from this time-consuming analytical task while maintaining policy accuracy

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system introduces an automated intermediary process that extracts content features, compares them against the security policy data store, and generates suggested security policies, mediating between the document content and the final security policy assignment

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11748501B2Tagging documents with security policies

Publication Date: 2023.09.05 ADOBE INC

AI summary

Embodiments of the present invention provide systems, methods, and computer storage media directed to facilitate identification of security policies for documents. In one embodiment, content features are identified from a set of documents having assigned security policies. The content features and corresponding security policies are analyzed to generate a security policy prediction model. Such a security policy prediction model can then be used to identify a security policy relevant to a document.