Tracked Without a Trace

Abstract

Behavior-based tracking is an unobtrusive technique that allows observers to monitor user activities on the Internet over long periods of time-in spite of changing IP addresses. Previous work has employed supervised classifiers in order to link the sessions of individual users. However, classifiers need labeled training sessions, which are difficult to obtain for observers. In this paper we show how this limitation can be overcome with an unsupervised learning technique. We present a modified k-means algorithm and evaluate it on a realistic dataset that contains the Domain Name System (DNS) queries of 3,862 users. For this purpose, we simulate an observer that tries to track all users, and an Internet Service Provider that assigns a different IP address to every user on every day. The highest tracking accuracy is achieved within the subgroup of highly active users. Almost all sessions of 73 % of the users in this subgroup can be linked over a period of 56 days. 19 % of the highly active users can be traced completely, i. e., all their sessions are assigned to a single cluster. This fraction increases to 40 % for shorter periods of seven days. As service providers may engage in behavior-based tracking to complement their existing profiling efforts, it constitutes a severe privacy threat for users of online services. Users can defend against behavior-based tracking by changing their IP address frequently, but this is cumbersome at the moment.

TransferGWAS: GWAS of images using deep transfer learning

“…We show that the learned representations are to some degree task-agnostic, meaning that even though the CNN has to be trained on a specific task (such as predicting disease states from images), the representations capture considerably more information than what is necessary to perform the specific transfer task. In the context of two-sample testing, a recent study has shown both theoretically and empirically that a similar transfer-hypothesis testing approach yields valid p-values and exhibits high statistical power [24].…”

The Wyner Variational Autoencoder for Unsupervised Multi-Layer Wireless Fingerprinting

“…For the baseline, in unsupervised learning setting, the stateof-the-art to the author's knowledge is the K-means based method [53]. We use an off-the-shelf K-means implementation [54], with input features formed from cascading the 200 × 2 bits traffic states and the 72 × 2 real value CSI (real and imaginary number as two independent channels).…”

“…Both methods use the two multi-layer features (traffic states and PHY CSI) without using the label information. The baseline follows[53].…”

On Cross-Layer Interactions of QUIC, Encrypted DNS and HTTP/3: Design, Evaluation, and Dataset

“…However, encrypted DNS has only recently gained traction with the standardisation of DNS over TLS (DoT) [3] and DNS over HTTPS (DoH) [4], where in today's Internet unencrypted DNS resolution using DNS over UDP (DoUDP) remains the default [5]. Hence, despite the encryption of the actual Web content, the browsing behaviors of individuals can still be observed, enabling third parties to create trackable user profiles [6][7][8][9].…”