t.ex-Graph: Automated Web Tracker Detection Using Centrality Metrics and Data Flow Characteristics

Philip Raschke, Patrick Herbke, Henry Schwerdtner

2023

Abstract

The practice of Web tracking raised concerns of privacy activists and data protection authorities over the past two decades. Simultaneously, researchers propose multiple solutions based on machine learning to automatically detect Web trackers. These solutions, while proving to be promising, often remained proofs-of-concept. This paper proposes t.ex-Graph, a representation that models data flows between websites to detect Web trackers in a graph. We use a publicly available dataset containing HTTP/S requests from a crawl of the Tranco top 10K websites to extract our graph. In the second step, we feed our graph into multiple machine-learning models to predict nodes that carry out tracking activities. Our results show high accuracy of 88% and even detect yet unknown Web trackers. We publish our artifacts for fellow researchers to replicate, reproduce, and advance our results.

Paper Citation

in Harvard Style

Raschke P., Herbke P. and Schwerdtner H. (2023). t.ex-Graph: Automated Web Tracker Detection Using Centrality Metrics and Data Flow Characteristics. In Proceedings of the 9th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP, ISBN 978-989-758-624-8, pages 199-209. DOI: 10.5220/0011787300003405

in Bibtex Style

@conference{icissp23,
author={Philip Raschke and Patrick Herbke and Henry Schwerdtner},
title={t.ex-Graph: Automated Web Tracker Detection Using Centrality Metrics and Data Flow Characteristics},
booktitle={Proceedings of the 9th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP,},
year={2023},
pages={199-209},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011787300003405},
isbn={978-989-758-624-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 9th International Conference on Information Systems Security and Privacy - Volume 1: ICISSP,
TI - t.ex-Graph: Automated Web Tracker Detection Using Centrality Metrics and Data Flow Characteristics
SN - 978-989-758-624-8
AU - Raschke P.
AU - Herbke P.
AU - Schwerdtner H.
PY - 2023
SP - 199
EP - 209
DO - 10.5220/0011787300003405