FP-Radar: Longitudinal Measurement and Early Detection of Browser\n Fingerprinting

Browser fingerprinting is a stateless tracking technique that attempts to\ncombine information exposed by multiple different web APIs to create a unique\nidentifier for tracking users across the web. Over the last decade, trackers\nhave abused several existing and newly proposed web APIs to further enhance the\nbrowser fingerprint. Existing approaches are limited to detecting a specific\nfingerprinting technique(s) at a particular point in time. Thus, they are\nunable to systematically detect novel fingerprinting techniques that abuse\ndifferent web APIs. In this paper, we propose FP-Radar, a machine learning\napproach that leverages longitudinal measurements of web API usage on top-100K\nwebsites over the last decade, for early detection of new and evolving browser\nfingerprinting techniques. The results show that FP-Radar is able to early\ndetect the abuse of newly introduced properties of already known (e.g., WebGL,\nSensor) and as well as previously unknown (e.g., Gamepad, Clipboard) APIs for\nbrowser fingerprinting. To the best of our knowledge, FP-Radar is also the\nfirst to detect the abuse of the Visibility API for ephemeral fingerprinting in\nthe wild.\n