Javier Parra-Arnau

Oualid Zari, Javier Parra-Arnau, Ayşe Ünsal et al.

In this paper, we present a stealthy and effective attack that exposes privacy vulnerabilities in Graph Neural Networks (GNNs) by inferring private links within graph-structured data. Focusing on the inductive setting where new nodes join the graph and an API is used to query predictions, we investigate the potential leakage of private edge information. We also propose methods to preserve privacy while maintaining model utility. Our attack demonstrates superior performance in inferring the links compared to the state of the art. Furthermore, we examine the application of differential privacy (DP) mechanisms to mitigate the impact of our proposed attack, we analyze the trade-off between privacy preservation and model utility. Our work highlights the privacy vulnerabilities inherent in GNNs, underscoring the importance of developing robust privacy-preserving mechanisms for their application.

Simon Hanisch, Patricia Arias-Cabarcos, Javier Parra-Arnau et al.

Our behavior (the way we talk, walk, act or think) is unique and can be used as a biometric trait. It also correlates with sensitive attributes like emotions and health conditions. Hence, techniques to protect individuals privacy against unwanted inferences are required, if such data is planned to be processed. To consolidate knowledge in this area, we systematically review applicable anonymization techniques. We taxonomize and compare existing solutions regarding privacy goals, conceptual operation, advantages, and limitations. We review anonymization techniques for the behavioral biometric traits of voice, gait, hand motions, eye-gaze, heartbeat (ECG), and brain activity (EEG). Our analysis shows that some behavioral traits (e.g., voice) have received much attention, while others (e.g., eye-gaze, brain activity) are mostly neglected. We also find that the evaluation methodology of behavioral anonymization techniques can be further improved.

Dominik Wunderlich, Daniel Bernau, Francesco Aldà et al.

Hierarchical text classification consists in classifying text documents into a hierarchy of classes and sub-classes. Although artificial neural networks have proved useful to perform this task, unfortunately they can leak training data information to adversaries due to training data memorization. Using differential privacy during model training can mitigate leakage attacks against trained models, enabling the models to be shared safely at the cost of reduced model accuracy. This work investigates the privacy-utility trade-off in hierarchical text classification with differential privacy guarantees, and identifies neural network architectures that offer superior trade-offs. To this end, we use a white-box membership inference attack to empirically assess the information leakage of three widely used neural network architectures. We show that large differential privacy parameters already suffice to completely mitigate membership inference attacks, thus resulting only in a moderate decrease in model utility. More specifically, for large datasets with long texts we observed Transformer-based models to achieve an overall favorable privacy-utility trade-off, while for smaller datasets with shorter texts convolutional neural networks are preferable.

Javier Parra-Arnau

Very recently, we are witnessing the emergence of a number of start-ups that enables individuals to sell their private data directly to brokers and businesses. While this new paradigm may shift the balance of power between individuals and companies that harvest data, it raises some practical, fundamental questions for users of these services: how they should decide which data must be vended and which data protected, and what a good deal is. In this work, we investigate a mechanism that aims at helping users address these questions. The investigated mechanism relies on a hard-privacy model and allows users to share partial or complete profile data with broker companies in exchange for an economic reward. The theoretical analysis of the trade-off between privacy and money posed by such mechanism is the object of this work. We adopt a generic measure of privacy although part of our analysis focuses on some important examples of Bregman divergences. We find a parametric solution to the problem of optimal exchange of privacy for money, and obtain a closed-form expression and characterize the trade-off between profile-disclosure risk and economic reward for several interesting cases.

Silvia Puglisi, Javier Parra-Arnau, Jordi Forné et al.

Recommendation systems and content filtering approaches based on annotations and ratings, essentially rely on users expressing their preferences and interests through their actions, in order to provide personalised content. This activity, in which users engage collectively, has been named social tagging. Although it has opened a myriad of new possibilities for application interoperability on the semantic web, it is also posing new privacy threats. Social tagging consists in describing online or online resources by using free-text labels (i.e. tags), therefore exposing the user's profile and activity to privacy attacks. Tag forgery is a privacy enhancing technology consisting of generating tags for categories or resources that do not reflect the user's actual preferences. By modifying their profile, tag forgery may have a negative impact on the quality of the recommendation system, thus protecting user privacy to a certain extent but at the expenses of utility loss. The impact of tag forgery on content-based recommendation is, therefore, investigated in a real-world application scenario where different forgery strategies are evaluated, and the consequent loss in utility is measured and compared.

Jagdish Prasad Achara, Javier Parra-Arnau, Claude Castelluccia

Free content and services on the Web are often supported by ads. However, with the proliferation of intrusive and privacy-invasive ads, a significant proportion of users have started to use ad blockers. As existing ad blockers are radical (they block all ads) and are not designed taking into account their economic impact, ad-based economic model of the Web is in danger today. In this paper, we target privacy-sensitive users and provide them with fine-grained control over tracking. Our working assumption is that some categories of web pages (for example, related to health, religion, etc.) are more privacy-sensitive to users than others (education, science, etc.). Therefore, our proposed approach consists in providing users with an option to specify the categories of web pages that are privacy-sensitive to them and block trackers present on such web pages only. As tracking is prevented by blocking network connections of third-party domains, we avoid not only tracking but also third-party ads. Since users will continue receiving ads on web pages belonging to non-sensitive categories, our approach essentially provides a trade-off between privacy and economy. To test the viability of our solution, we implemented it as a Google Chrome extension, named MyTrackingChoices (available on Chrome Web Store). Our real-world experiments with MyTrackingChoices show that the economic impact of ad blocking exerted by privacy-sensitive users can be significantly reduced.

Javier Parra-Arnau, Jagdish Prasad Achara, Claude Castelluccia

The intrusiveness and the increasing invasiveness of online advertising have, in the last few years, raised serious concerns regarding user privacy and Web usability. As a reaction to these concerns, we have witnessed the emergence of a myriad of ad-blocking and anti-tracking tools, whose aim is to return control to users over advertising. The problem with these technologies, however, is that they are extremely limited and radical in their approach: users can only choose either to block or allow all ads. With around 200 million people regularly using these tools, the economic model of the Web ---in which users get content free in return for allowing advertisers to show them ads--- is at serious peril. In this paper, we propose a smart Web technology that aims at bringing transparency to online advertising, so that users can make an informed and equitable decision regarding ad blocking. The proposed technology is implemented as a Web-browser extension and enables users to exert fine-grained control over advertising, thus providing them with certain guarantees in terms of privacy and browsing experience, while preserving the Internet economic model. Experimental results in a real environment demonstrate the suitability and feasibility of our approach, and provide preliminary findings on behavioral targeting from real user browsing profiles.