Web Tracker Classification Using Obfuscated Metrics and ML
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing web tracking prevention systems face challenges in accurately identifying and mitigating web trackers while preserving user privacy and maintaining compatibility with important websites, as trackers often evade detection through domain changes and localized delivery points, leading to potential privacy breaches and compatibility issues.
Innovation Solution
A system that intelligently classifies web trackers using obfuscated metrics scaled to user browsing history and unknown quantities, employing a machine learning model to determine tracker classifications and selectively relax restrictions based on user engagement levels, ensuring privacy preservation and compatibility with frequently visited sites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If storage access restrictions are imposed on web trackers, then user privacy is protected, but compatibility issues arise with sites that expect unrestricted tracker operation
Solution Approach 1:
The system applies different storage access policies to different web trackers based on their classification. High-value trackers (those associated with important websites) are permitted greater storage access, while low-value trackers face stricter restrictions. This local differentiation resolves the contradiction by tailoring the strictness of privacy protection to the specific tracker and website context.
Solution Approach 2:
The system dynamically adjusts storage access restrictions based on real-time classification results and user engagement metrics. Rather than applying static restrictions to all trackers, the system continuously monitors tracker behavior and website importance, adjusting access permissions accordingly. This dynamic approach maintains privacy protection while ensuring compatibility with important sites.
2Measurement precision
If tracking prevention systems use known tracker signatures, then detection accuracy is maintained, but trackers evade detection by changing domain names and delivery points
Solution Approach 1:
The system performs preliminary classification of web trackers using machine learning models before applying storage access restrictions. By analyzing tracker characteristics, behavior patterns, and website context in advance, the system identifies high-value trackers that should be permitted. This preliminary action enables accurate detection of legitimate trackers while maintaining privacy protection against malicious ones.
Solution Approach 2:
The system introduces an intermediary classification layer between tracker detection and storage access restriction. Rather than directly blocking all trackers or permitting all access, the classification system acts as an intermediary that evaluates tracker legitimacy and mediates appropriate access levels. This intermediary mechanism maintains detection accuracy while adapting to tracker evasion techniques.
3Reliability
If web trackers are restricted from storing per-user identifiers, then user tracking is prevented, but user engagement with important websites deteriorates
Solution Approach 1:
The system applies different storage access policies to different trackers based on website importance and user engagement metrics. For high-value trackers associated with important websites, the system permits greater storage access including per-user identifier storage, while maintaining strict restrictions on low-value trackers. This local differentiation preserves anti-tracking effectiveness while maintaining user experience for important sites.
Solution Approach 2:
The system uses feedback from user engagement metrics and website importance assessments to adjust storage access permissions. By monitoring user interactions and website significance, the system learns which trackers are valuable for user experience and adjusts restrictions accordingly. This feedback mechanism ensures anti-tracking effectiveness is maintained while preventing deterioration of user experience.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Embodiments described herein are directed to intelligently classifying Web trackers in a privacy preserving manner and mitigating the effects of such Web trackers. As users browse the Web and encounter various Web sites, tracker-related metrics are determined. The metrics are obfuscated to protect the privacy of the user. The obfuscated metrics are provided as inputs to a machine learning model, which is configured to output a classification for the Web trackers associated with the Web sites visited by the user. Depending on the classification, the effects of the Web trackers are mitigated by placing restrictions on the Web trackers. The restrictions for a particular Web tracker may be relaxed based on a level of user engagement a user has with respect to the tracker's associated Web site. By doing so, the compatibility risks associated with tracking prevention are mitigated for Web sites that are relatively important to the user.