AutoFR: Automated Filter Rule Generation for Adblocking

Abstract

Adblocking relies on filter lists, which are manually curated and maintained by a small community of filter list authors. This manual process is laborious and does not scale well to a large number of sites and over time. We introduce Aut-oFR, a reinforcement learning framework to fully automate the process of filter rule creation and evaluation. We design an algorithm based on multi-arm bandits to generate filter rules while controlling the trade-off between blocking ads and avoiding breakage. We test our implementation of AutoFR on thousands of sites in terms of efficiency and effectiveness. AutoFR is efficient: it takes only a few minutes to generate filter rules for a site. AutoFR is also effective: it generates filter rules that can block 86% of the ads, as compared to 87% by EasyList while achieving comparable visual breakage. The filter rules generated by AutoFR generalize well to new and unseen sites. We envision AutoFR to assist the adblocking community in automated filter rule generation at scale.

Blocking JavaScript without Breaking the Web: An Empirical Investigation

“…Functionality breakage is a subjective metric that requires visual inspection, which is not feasible on 100K webpages. We discuss two metrics that can be automatically calculated and are correlated with website breakage [53]. Tracking and Functional request count: Network requests fetch critical functional resources like scripts, images, and other media as well as JS scripts and images that perform tracking activity.…”

“…The typical evasion strategy is to manipulate the URLs, e.g., change the URL path or hostname, such that filter lists are no longer effective [25,40]. This has led to an arms race where filter lists need to be promptly updated in response to such evasion attempts [30,53,71]. Filter list curators have also made a concerted effort to selectively block the underlying scripts from downloading or execution that are responsible for initiating tracking requests.…”