AI Governance System for Confidential Data Anonymization

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional systems for identifying and protecting confidential information in data networks are time and resource intensive, and they may not detect and secure all necessary confidential information.

Innovation Solution

A comprehensive artificial intelligence-enabled governance framework that processes textual and document data, including image data, to identify and anonymize confidential information using optimized machine learning models, reducing resource reliance and enhancing efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional systems employ manual and automated techniques for identifying confidential information, then the system can detect and protect confidential data, but the process becomes time and resource intensive

Engineering Contradiction:

Improvedetection of confidential informationVSAvoidtime consumption

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces manual inspection processes with automated machine learning models that can rapidly analyze text data to identify confidential information. The system uses trained models to automatically detect patterns, entities, and contexts that indicate confidential data, eliminating the need for manual review while maintaining high detection accuracy.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the approach from traditional rule-based detection to machine learning-based detection, utilizing parameters such as text embeddings, entity relationships, and contextual patterns. This allows the system to adapt to various forms of confidential information and improve detection capabilities over time without increasing time consumption.

Inventive Principle:

Principle #35Parameter changes

2Productivity

If conventional systems employ automated data monitoring techniques, then the system can analyze network data to determine unauthorized activity, but the systems are not guaranteed to detect and protect all necessary confidential information

Engineering Contradiction:

Improveautomated data monitoringVSAvoiddetection completeness

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent segments the data analysis process into multiple stages: extracting text from various formats, identifying entities and relationships, determining confidentiality based on contextual analysis, and generating protection actions. This segmentation allows each component to specialize in specific tasks, improving overall detection completeness while maintaining productivity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary layer of machine learning models that act as mediators between raw data and protection decisions. These models analyze text data, identify confidential patterns, and provide recommendations for protection actions, enabling more comprehensive and accurate detection compared to traditional automated monitoring.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If the system processes various types of data including text and document data with image data, then the system can comprehensively identify confidential information, but the processing complexity increases

Engineering Contradiction:

Improveprocessing capabilityVSAvoidprocessing complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a universal processing framework that can handle multiple data types (text, documents, images) through a common machine learning pipeline. The system uses a unified approach where text extraction, entity identification, and confidentiality determination are performed across all data types, reducing the need for separate specialized systems for each data format.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250165650A1System and method for performing artificial intelligence governance

Publication Date: 2025.05.22 POLYGRAF INC

AI summary

A method for identifying and anonymizing confidential information in input data where the method is performed by at least one computer processor executing computer-readable instructions tangibly stored on at least one computer-readable medium. The method includes extracting one or more of textual data and document data from the input data, wherein the document data includes one or more of second textual data and image data, processing the textual data and the second textual data to identify the confidential information therein, anonymizing the confidential information, processing the image data to identify image-based confidential information therein, and anonymizing the image-based confidential information.