Knowledge Graph Verification for Data Leakage Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies face challenges in effectively detecting data leakage and dangerous information, particularly in identifying whether statements are public or private and true or false, which is crucial for information security and user safety.
Innovation Solution
A computer-implemented method utilizing a knowledge graph to detect data leakage and dangerous information by receiving a knowledge graph, extracting data from network services, identifying statements, and determining their public or private status and truthfulness using the knowledge graph.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional data leakage detection methods (watermarking, pattern matching) are used, then data leakage can be detected, but the ability to determine truthfulness and public/private status of statements is insufficient
Solution Approach 1:
The patent introduces a knowledge graph as an intermediary system that mediates between extracted statements and verification processes. The knowledge graph stores structured information about entities, relationships, and metadata (including public/private status and truthfulness), enabling comprehensive verification of statements without directly analyzing unstructured text for these attributes
Solution Approach 2:
The patent creates a copy of the knowledge graph specifically for verification purposes. This copy contains structured representations of statements with associated metadata about public/private status and truthfulness, allowing the system to verify statements against the knowledge graph without modifying the original knowledge graph structure
2Reliability
If language models are used to detect data leakage and dangerous information, then detection capability is enhanced, but the system complexity and computational resources required increase significantly
Solution Approach 1:
The patent segments the detection system into distinct functional components: data extraction module, knowledge graph module, and verification module. Each component handles specific tasks independently - extraction retrieves data from network services, the knowledge graph stores and structures information, and verification determines public/private status and truthfulness, reducing overall system complexity
Solution Approach 2:
The knowledge graph serves multiple functions simultaneously: it stores entity information, maintains relationship data, tracks public/private status metadata, and provides verification capabilities. This multi-functionality reduces the need for separate specialized systems, thereby reducing overall complexity
3Reliability
If comprehensive verification of statement public/private status and truthfulness is implemented, then information security is improved, but processing time and computational resources increase
Solution Approach 1:
The system performs preliminary actions by pre-processing and storing verification information in the knowledge graph before actual verification is needed. Metadata about public/private status and truthfulness is pre-established and stored, allowing rapid retrieval and verification without time-consuming analysis during the verification process
Solution Approach 2:
The patent uses a copy of the knowledge graph for verification operations, allowing the original knowledge graph to remain intact while the copy is used for time-sensitive verification tasks. This enables parallel processing and reduces the time required for verification by avoiding modifications to the original structure
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Aspects relate to a computer implemented method, a computer program and a computer system for detecting data leakage and/or detecting dangerous information. The method comprises receiving a knowledge graph and extracting data from at least one network service. The method further comprises identifying statements in the extracted data. For each identified statement, the method further comprises determining whether the identified statement is public or private using the knowledge graph, and/or determining whether the identified statement is true or false using the knowledge graph.