Semantic Watermarking for Unique File Clone Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Information Rights Management (IRM) solutions fail to prevent file leaks through screenshotting or manual copying of documents, even when access is legitimate, and lack user-specific identification of leak sources, which hampers deterrent and recovery efforts.
Innovation Solution
A method involving semantic differencing elements is used to generate unique clone copies of files for each user, incorporating changes like punctuation marks, synonyms, and formatting adjustments, with a difference score system to control access and identify leaks by comparing recovered files to stored watermarks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional watermarking solutions are used to identify confidential information, then confidentiality indication is improved, but user-specific leak source identification capability deteriorates
Solution Approach 1:
The patent segments the watermarking approach by creating multiple unique watermarks for different users rather than using a single generic watermark. Each user receives a customized version of the document with their specific watermark embedded, enabling precise identification of leak sources while maintaining confidentiality indication.
Solution Approach 2:
The patent applies local quality by making each watermark unique to a specific user context. Instead of uniform watermarking across all users, the system embeds location-specific, user-specific watermarks that provide both confidentiality indication and precise leak source identification tailored to each user's access context.
2Reliability
If encryption and access controls are implemented to protect files, then unauthorized access prevention is improved, but protection against screenshotting and manual copying deteriorates
Solution Approach 1:
The patent applies preliminary action by embedding watermarks into documents before distribution to users. This pre-embedding ensures that even if users screenshot or manually copy content, the watermark information is already present in the original document structure, enabling later leak source identification despite the limitations of encryption and access controls.
Solution Approach 2:
The patent introduces watermarks as an intermediary mechanism that operates independently of encryption and access controls. These watermarks serve as a separate layer of protection that can identify leak sources even when traditional security measures fail to prevent screenshotting or manual copying, acting as a mediator between access control and leak prevention.
3Loss of information
If generic watermarks are used to identify confidential documents, then confidentiality marking is improved, but user-specific identification and deterrent effect deteriorates
Solution Approach 1:
The patent implements local quality by customizing watermarks for each user based on their specific access context, document, and user profile. This user-specific approach enhances both the identification capability (by uniquely linking leaks to specific users) and the deterrent effect (by making each user aware their specific actions are traceable), while maintaining confidentiality marking functionality.
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting watermark characteristics based on user-specific parameters such as user ID, document ID, timestamp, and access context. This creates adaptable watermarks that provide user-specific identification and enhanced deterrent effects while maintaining the core confidentiality marking function.
Data Source
AI summary
Methods and systems for identifying a source of leak of confidential information are described herein. The methods and systems related to a file sharing system that may generate a copy of a shared file that is unique each user. The file management system may determine a plurality of differences within each version of the file. A table of differences is maintained by the file management system, for comparison against any leaked version of the file. The file management system compares each generated version of the file against previously generated versions to ensure enough differences are included to determine a unique identity of the user associated with any leaked file.