A tracking flow is a between an end user and Web service. We develop extensive measurement methodology for quantifying at scale the amount of flows that cross data protection borders, be it national or international, such as EU28 border within which General Data Protection Regulation (GDPR) applies. Our uses browser extension to fully render advertising code, various lists heuristics extract well known trackers, passive DNS replication get all IP ranges state-of-the art geolocation. employ...

Real networks are complex dynamical systems, evolving over time with the addition and deletion of nodes links. Currently, there exists no principled mathematical theory for their dynamics-a grand-challenge open problem. Here, we show that popularity similarity trajectories in hyperbolic embeddings different real manifest universal self-similar properties typical Hurst exponents H≪0.5. This means predictable, displaying antipersistent or "mean-reverting" behavior, they can be adequately...

We present a Federated Learning (FL) based solution for building distributed classifier capable of detecting URLs containing sensitive content, i.e., content related to categories such as health, political beliefs, sexual orientation, etc.Although addresses the limitations previous offline/centralised classifiers, it is still vulnerable poisoning attacks from malicious users that may attempt reduce accuracy benign by disseminating faulty model updates.To guard against this, we develop robust...

In recent years, governments worldwide have moved their services online to better serve citizens. Benefits aside, this choice increases the danger of tracking via such sites. This is great concern as governmental websites increasingly become only interaction point with government. paper, we investigate popular across different countries and assess what extent visits these sites are tracked by third-parties. Our results show that, unfortunately, a serious concern, in some up 90% create...

Networks-on-Chips (NoCs) are experiencing escalating susceptibility to wear-out and reduced reliability, with the risk of becoming key point failure in an entire multicore chip. Aiming towards seamless NoC operation presence faulty communication links, this paper we propose Hermes, a highly-robust, distributed lightweight fault-tolerant routing algorithm, whose performance degrades gracefully increasing link counts. Hermes is deadlock-free hybrid utilizing load-balancing on fault-free paths...

We present the design, implementation, validation, and deployment of Price Sheriff, a highly distributed system for detecting various types online price discrimination in e-commerce. The Sheriff uses peer-to-peer architecture, sandboxing, secure multiparty computation to allow users tunnel check requests through browsers other peers without tainting their local or server-side browsing history state. Having operated several months with approximately one thousand real users, we identify...

Online Behavioral Advertising (OBA) is an important revenue source for online publishers and content providers. However, the extensive user tracking required to enable OBA raises valid privacy concerns. Existing proposed solutions either block all tracking, therefore breaking entirely, or require significant changes on current advertising infrastructure, making adoption hard. We propose Web Identity Translator (WIT), a new service running as proxy middlebox. WIT stops original cookies from...

This paper aims to stir debate about a disconcerting privacy issue on web browsing that could easily emerge because of unethical practices and uncontrolled use technology. We demonstrate how straightforward is capture behavioral data the users at scale, by unobtrusively tracking their mouse cursor movements, predict user's demographics information with reasonable accuracy using five lines code. Based our results, we propose an adversarial method mitigate user profiling techniques make...

Several data protection laws include special provisions for protecting personal relating to religion, health, sexual orientation, and other sensitive categories. Having a well-defined list of categories is sufficient filing complaints manually, conducting investigations, prosecuting cases in courts law. Data laws, however, do not define explicitly what type content falls under each category. Therefore, it unclear how implement proactive measures such as informing users, blocking trackers,...

A large number of Data Marketplaces (DMs) have appeared in the last few years to help owners monetize their data, and data buyers optimize marketing campaigns, train ML models, facilitate other data-driven decision processes. In this paper, we present a first its kind measurement study growing DM ecosystem, focused on understanding which features are actually driving prices market. We show that products listed commercial DMs may cost from hundreds thousands US dollars. analyze different...

A large number of Data Marketplaces (DMs) have appeared in the last few years to help owners monetise their data, and data buyers optimize marketing campaigns, train ML models, facilitate other data-driven decision processes. In this paper, we present a first its kind measurement study growing DM ecosystem, shedding light on several totally unknown facts about it. We show that products listed commercial DMs may cost from hundreds thousands US dollars. analyse prices different categories...

As recent events have demonstrated, disinformation spread through social networks can dire political, economic and consequences. Detecting must inevitably rely on the structure of network, users particularities event occurrence patterns. We present a graph data structure, which we denote as meta-graph, that combines underlying users' relational information, well semantic topical modeling. detail construction an example meta-graph using Twitter covering 2016 US election campaign then compare...

What if instead of having to implement controversial user tracking techniques, Internet advertising & marketing companies asked explicitly be granted access data by name and category, such as Alice→Mobility→05-11-2020? The technology for implementing this already exists, is none other than the Information Centric Networks (ICN), developed over a decade in framework Next Generation (NGI) initiatives. Beyond named personal data, ICN's in-network storage capability can used substrate...

Networks-on-Chips (NoCs) are experiencing escalating susceptibility to wear-out and reduced reliability, with the risk of becoming key point failure in an entire multicore chip. In this paper we propose Hermes, a highly-robust, distributed fault-tolerant routing algorithm, whose performance degrades gracefully increasing faulty NoC link counts. Hermes is deadlock-free hybrid utilizing load-balanced on fault-free paths, while providing pre-reconfigured escape routes vicinity faults. An...

Targeted online advertising is a multi-billion dollar business based on the ability of profiling and delivering targeted ads to wide range users. Due privacy erosion associated with such business, researchers are trying understand how works anti-tracking applications becoming popular among Both research privacy-enhancing apps, however, target ad-networks or over-the-top providers that have unrestricted access users' activity. There seems be little interest in potential activities by "network...

We turn our attention to the elephant in room of data protection, which is none other than simple and obvious question: "Who's tracking sensitive domains?". Despite a fast-growing amount work on more complex facets interplay between privacy business models Web, question who collects domains where most people would prefer not be seen, has received rather limited attention. First, we develop methodology for automatically annotating websites that belong category, e.g. as defined by General Data...

Over the years, web content has evolved from simple text and static images hosted on a single server to complex, interactive multimedia-rich different servers. As result, modern website during its loading time fetches not only owner's domain but also range of third-party domains providing additional functionalities services. Here, we infer network by observing domains' interactions within users' browsers all over globe. We find that this possesses structural properties commonly found in...

With relentless CMOS technology downsizing Networks-on-Chips (NoCs) are inescapably experiencing escalating susceptibility to wearout and reduced reliability. While faults in processors memories may be masked via redundancy, or mitigated techniques such as task migration, NoCs especially vulnerable hardware a single link breakdown cause inter-tile communication halt indefinitely, rendering the whole multicore chip inoperable. As such, impose risk of becoming pivotal point failure multicores...

It is a widely accepted fact that state-sponsored Twitter accounts operated during the 2016 US presidential election, spreading millions of tweets with misinformation and inflammatory political content. Whether these social media campaigns so-called "troll" were able to manipulate public opinion still in question. Here, we quantify influence troll on by analyzing 152.5 million (by 9.9 users) from period. The data contain original 822 identified as such itself. We construct analyse very large...

A large number of Data Marketplaces (DMs) have appeared in the last few years to help owners monetise their data, and data buyers fuel marketing process, train ML models, perform other data-driven decision processes. In this paper, we present a first its kind measurement study growing DM ecosystem shed light on several totally unknown facts about it. For example, show that median price live products sold under subscription model is around US\$1,400 per month. one-off purchases static...