Study Reveals Gaps in Browser Fingerprinting Research
/ 1 min read
🕵️♂️ New study reveals significant gaps in browser fingerprinting research. A recent user study involving 30 participants over 10 weeks found that automated web crawls miss nearly 45% of fingerprinting websites encountered by real users. This discrepancy arises from crawlers’ limitations in accessing authentication-protected pages and triggering fingerprinting scripts that require specific user interactions. The research highlights new fingerprinting vectors identified in real user data that were absent in automated crawls. Additionally, the study demonstrates that federated learning can enhance the performance of browser fingerprinting detection models when trained on real user data, outperforming those trained solely on automated crawl data. This underscores the need for more accurate methods to understand online tracking techniques.
