Detecting Label Data Leakage Channels via Dynamic Watermarking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional database watermarking and data trajectory tracking technologies are ineffective in detecting and tracking user label data due to its non-numeric nature, dispersed usage, and dynamic changes, making it difficult to identify data leakage channels in uncontrollable partner environments.
Innovation Solution
A method and apparatus for detecting label data leakage channels by adding detection labels to user data sets, establishing channel indices, intercepting and analyzing push information based on probabilities, and updating labels to identify suspected leakage channels, utilizing a HASH function for credibility calculation and sampling to ensure low co-occurrence with existing labels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional database watermarking technology is used, then data can be tracked in controlled environments, but it cannot effectively detect label data leakage because user labels do not include numeric fields and are used in a dispersed manner
Solution Approach 1:
The patent transforms non-numeric label data into detectable forms by computing co-occurrence probabilities between labels and converting them into numerical watermark signals. This parameter transformation enables conventional watermarking techniques to detect leakage in non-numeric label data while maintaining their tracking capabilities.
2Loss of information
If label data is used in a dispersed manner for user preferences, then user privacy is protected, but it becomes difficult to detect watermarks and track data leakage channels
Solution Approach 1:
The system establishes feedback mechanisms by monitoring push information and advertising data to detect whether watermarked label data has been leaked. By continuously observing data usage patterns and comparing them against the embedded watermarks, the system can identify leakage channels while preserving the dispersed nature of label data for privacy protection.
3Reliability
If conventional data trajectory tracking technology is used, then data flow can be monitored in controlled environments, but it cannot address the challenges of massive and dynamic user label data that changes over time
Solution Approach 1:
The patent embeds watermarks into label data before it leaves the controlled environment, performing preliminary marking actions. This allows the system to track massive and dynamic label data efficiently without requiring real-time analysis of all data flows, as the watermarks are already in place to identify leakage channels when push information is received.
4Reliability
If detection labels are added to user label data sets, then data leakage channels can be detected, but the complexity of the detection system increases
Solution Approach 1:
The detection labels serve multiple functions: they act as watermarks embedded in label data, serve as identifiers for tracking data flows, and function as keys for detecting leakage channels. This multi-functionality reduces the need for separate detection mechanisms, thereby managing system complexity while maintaining high detection accuracy.
Data Source
AI summary
The present disclosure provides label data leakage channel detection methods and apparatuses. According to one exemplary label data leakage channel detection method, detection labels are generated based on normal labels of a user. The detection labels can be associated with different data usage channels, so as to indirectly detect usage of the detection labels. Possible data leakage channels can be effectively detected based on massive data indexing and searching. One exemplary apparatus of the present disclosure includes a detection label adding module, a channel association module, an interception module, an intercepted information analysis module, a channel searching module, and an output module. The detection methods and apparatuses provided by the present disclosure have the advantages of high detection efficiency and the capability of processing massive and dynamic user label data.