Automated Data Masking with False Positive Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing automated tools for masking sensitive data often result in false positives, leading to inaccurate masking and the need for manual correction, which consumes computing and networking resources.

Innovation Solution

A system that includes a detection model for identifying potential sensitive data fields using data type matching and pattern matching, and a false positive model using contextual analysis to differentiate between sensitive and non-sensitive data fields, thereby masking only the sensitive fields and avoiding false positives.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated tools are used to mask sensitive data, then masking speed and security are improved, but false positives increase leading to inaccurate masking

Engineering Contradiction:

Improvemasking speedVSAvoidmasking accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The masking system is divided into multiple specialized components: a detection model for identifying potential sensitive data fields using data type matching and pattern matching, and a separate false positive model for verifying contextual accuracy. This segmentation allows each component to specialize in its function, improving both speed and accuracy independently.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

A false positive detection model acts as an intermediary between the initial detection model and the final masking operation. This intermediary layer verifies potential matches using contextual analysis, eliminating false positives before they result in inaccurate masking, thereby maintaining both high speed and high accuracy.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If manual correction is performed to fix false positives, then masking accuracy is improved, but computational resources and time are consumed

Engineering Contradiction:

Improvemasking accuracyVSAvoidcorrection time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs self-correction through automated false positive detection and avoidance mechanisms. The false positive model automatically identifies and eliminates erroneous detections without requiring manual intervention, maintaining high accuracy while eliminating the time loss associated with manual correction.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The false positive detection model performs preliminary verification before the final masking operation. By proactively identifying and eliminating false positives in advance, the system prevents inaccurate masking without requiring subsequent manual correction, saving both time and computational resources.

Inventive Principle:

Principle #10Preliminary action

3Reliability

If comprehensive pattern matching is used to identify sensitive data, then detection coverage is improved, but false positive rate increases

Engineering Contradiction:

Improvedetection coverageVSAvoidfalse positive rate

Core Design Contradiction:

ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The detection process is segmented into two stages: first, the detection model uses comprehensive pattern matching to ensure high detection coverage; second, the false positive model applies contextual analysis to filter out false positives. This segmentation allows comprehensive matching without being penalized by the resulting false positives.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The false positive model serves as an intermediary filtering layer that receives comprehensive detection results from the pattern matching model and eliminates false positives through contextual verification. This allows the system to maintain high detection coverage while reducing the false positive rate.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12321496B2Automated data masking with false positive detection and avoidance

Publication Date: 2025.06.03 CAPITAL ONE SERVICES LLC

AI summary

In some implementations, a system may receive a set of data intended for storage. The system may detect, within the set of data and using pattern matching, a set of potential sensitive data fields. The system may detect, using characters not included in the potential sensitive data fields, at least one non-sensitive data field included in the set of potential sensitive data fields. The system may mask first data included in the set of potential sensitive data fields other than the at least one non-sensitive data field and may refrain from masking second data included in the at least one non-sensitive data field. The system may generate a modified set of data based on masking the first data and refraining from masking the second data and output the modified set of data for storage.