AdGraph: A Graph-Based Approach to Ad and Tracker Blocking

Abstract

User demand for blocking advertising and tracking online is large and growing. Existing tools, both deployed and described in research, have proven useful, but lack either the completeness or robustness needed for a general solution. Existing detection approaches generally focus on only one aspect of advertising or tracking (e.g. URL patterns, code structure), making existing approaches susceptible to evasion.In this work we present ADGRAPH, a novel graph-based machine learning approach for detecting advertising and tracking resources on the web. ADGRAPH differs from existing approaches by building a graph representation of the HTML structure, network requests, and JavaScript behavior of a webpage, and using this unique representation to train a classifier for identifying advertising and tracking resources. Because ADGRAPH considers many aspects of the context a network request takes place in, it is less susceptible to the single-factor evasion techniques that flummox existing approaches.We evaluate ADGRAPH on the Alexa top-10K websites, and find that it is highly accurate, able to replicate the labels of human-generated filter lists with 95.33% accuracy, and can even identify many mistakes in filter lists. We implement ADGRAPH as a modification to Chromium. ADGRAPH adds only minor overhead to page loading and execution, and is actually faster than stock Chromium on 42% of websites and AdBlock Plus on 78% of websites. Overall, we conclude that ADGRAPH is both accurate enough and performant enough for online use, breaking comparable or fewer websites than popular filter list based approaches.

Tales of Favicons and Caches: Persistent Tracking in Modern Browsers

“…Englehardt and Narayanan [25] conducted a large scale measurement study to quantify the use of stateful and stateless tracking and cookie syncing. Numerous studies have also proposed techniques for blocking trackers [39], [35], [38], [86]. On the other hand, out paper demonstrates a novel technique that allows websites to re-identify users.…”

Privacy in targeted advertising: A survey

“…direct and inferred leakage), have been examined extensively in literature, e.g. [10], [97], including third party ad tracking and visiting [98], [99].…”

URL-based Web Tracking Detection Using Deep Learning

“…Regarding tracking detection methods, Ikram et al in [9] and Wu et al in [8] proposed machine learning algorithms based on the analysis of JavaScript code syntax to detect tracking patterns. Other works, such as Iqbal et al in [7], proposed complex combinations of JavaScript code attribution, DOM code inspection and network requests annotation to detect tracking pattern signatures. However, most of these approaches are very specialized and complex, and usually require major modifications to the browsers, fact that prevent common users to adopt these solutions.…”

“…Although the research community has dedicated great efforts to improve the privacy protection measures by means of new complex techniques (e.g. [7]- [9]), the most popular approach to block web tracking systems still relies on traditional content-blockers [10]; browser plugins that block URLs found on manually-curated pattern lists. The reason behind this is that latest methods proposed in the literature are difficult to adopt by common users.…”

“…The reason behind this is that latest methods proposed in the literature are difficult to adopt by common users. For example, recent proposals require users to install instrumented versions of web browsers [7], get measurements from multiple vantage points [11] or share privacy-sensitive information, like cookies [12].…”