arXiv subjects

Imane Fouad

Publications and source records attributed to Imane Fouad.

6 recordsLinked to original sources

arXiv2026

Extracting and Verifying Illicit Bitcoin Addresses from Underground Forum Discussions

Existing labeled Bitcoin datasets are largely derived from community-reported abuse, blockchain heuristics, incident-specific collections, or proprietary labeling processes. Their construction methods are rarely publicly reproducible and often provide limited evidence that an address was directly involved in illicit activity. We present a reproducible pipeline for constructing evidence-backed Bitcoin labels from HackForums, an underground cybercrime forum with fifteen years of archived activity. The pipeline combines LLM-assisted screening, expert review, and on-chain validation to identify Bitcoin addresses explicitly associated with illicit transactions discussed on the forum. Each released label is supported by contextual evidence from underground discussions and validated on-chain. The resulting dataset contains 2,438 manually verified illicit Bitcoin addresses spanning 2010-2024 and twelve cybercrime categories assigned during LLM screening. We release the dataset, temporal metadata, and the complete extraction pipeline to support reproducible research on cryptocurrency-facilitated cybercrime.

cs.CR↗

arXiv2026

Can We Unmask the Underground? Detecting and Predicting Hidden Forum Interactions

Cybercriminal underground forums enable anonymous collaboration, allowing users to trade illicit tools, discuss vulnerabilities, and distribute stolen data. Driven by shared interests and specialized skills, users on these platforms organize into distinct threat communities. However, identifying these communities is challenging due to their dynamic and opaque structures; traditional graph-based methods typically isolate dominant groups while overlooking smaller, hidden subgroups. This paper introduces HADES, an unsupervised framework designed to detect both dominant and hidden communities in underground forums. The framework models users based on their textual interactions and leverages pretrained language models to generate semantic embeddings that encode latent behavioral and thematic patterns. By clustering users based on semantic similarity and assigning topic labels to the resulting clusters, HADES identifies specific threat communities to support Cyber Threat Intelligence (CTI). The framework was evaluated on three major underground forums: HackForums, Cracked, and BreachForums. Results demonstrate that BERT embeddings consistently outperform alternative baselines, improving cluster coherence and achieving higher silhouette scores. Across these platforms, HADES identified dozens of distinct communities, effectively isolating small, specialized subgroups (fewer than 100 users) that evade traditional detection. Furthermore, because shared thematic interests frequently precede explicit structural connections, tracking these semantic patterns enables the framework to anticipate the formation of threat communities up to a year before they become detectable by traditional graph-based methods.

cs.SI↗

arXiv2025

FakeZero: Real-Time, Privacy-Preserving Misinformation Detection for Facebook and X

Social platforms distribute information at unprecedented speed, which in turn accelerates the spread of misinformation and threatens public discourse. We present FakeZero, a fully client-side, cross-platform browser extension that flags unreliable posts on Facebook and X (formerly Twitter) while the user scrolls. All computation, DOM scraping, tokenization, Transformer inference, and UI rendering run locally through the Chromium messaging API, so no personal data leaves the device. FakeZero employs a three-stage training curriculum: baseline fine-tuning and domain-adaptive training enhanced with focal loss, adversarial augmentation, and post-training quantization. Evaluated on a dataset of 239,000 posts, the DistilBERT-Quant model (67.6 MB) reaches 97.1% macro-F1, 97.4% accuracy, and an AUROC of 0.996, with a median latency of approximately 103 ms on a commodity laptop. A memory-efficient TinyBERT-Quant variant retains 95.7% macro-F1 and 96.1% accuracy while shrinking the model to 14.7 MB and lowering latency to approximately 40 ms, showing that high-quality fake-news detection is feasible under tight resource budgets with only modest performance loss. By providing inline credibility cues, the extension can serve as a valuable tool for policymakers seeking to curb the spread of misinformation across social networks. With user consent, FakeZero also opens the door for researchers to collect large-scale datasets of fake news in the wild, enabling deeper analysis and the development of more robust detection techniques.

cs.CR↗

arXiv2023

UA-Radar: Exploring the Impact of User Agents on the Web

In the early days of the web, giving the same web page to different browsers could provide very different results. As the rendering engine behind each browser would differ, some elements of a page could break or be positioned in the wrong location. At that time, the User Agent (UA) string was introduced for content negotiation. By knowing the browser used to connect to the server, a developer could provide a web page that was tailored for that specific browser to remove any usability problems. Over the past three decades, the UA string remained exposed by browsers, but its current usefulness is being debated. Browsers now adopt the exact same standards and use the same languages to display the same content to users, bringing the question if the content of the UA string is still relevant today, or if it is a relic of the past. Moreover, the diversity of means to browse the web has become so large that the UA string is one of the top contributors to tracking users in the field of browser fingerprinting, bringing a sense of urgency to deprecate it. In this paper, our goal is to understand the impact of the UA on the web and if this legacy string is still actively used to adapt the content served to users. We introduce UA-Radar, a web page similarity measurement tool that compares in-depth two web pages from the code to their actual rendering, and highlights the similarities it finds. We crawled 270, 048 web pages from 11, 252 domains using 3 different browsers and 2 different UA strings to observe that 100% of the web pages were similar before any JavaScript was executed, demonstrating the absence of differential serving. Our experiments also show that only a very small number of websites are affected by the lack of UA information, which can be fixed in most cases by updating code to become browser-agnostic. Our study brings some proof that it may be time to turn the page on the UA string and retire it from current web browsers.

cs.CR↗

arXiv2021

Did I delete my cookies? Cookies respawning with browser fingerprinting

Stateful and stateless web tracking gathered much attention in the last decade, however they were always measured separately. To the best of our knowledge, our study is the first to detect and measure cookie respawning with browser and machine fingerprinting. We develop a detection methodology that allows us to detect cookies dependency on browser and machine features. Our results show that 1,150 out of the top 30, 000 Alexa websites deploy this tracking mechanism. We further uncover how domains collaborate to respawn cookies through fingerprinting. We find out that this technique can be used to track users across websites even when third-party cookies are deprecated. Together with a legal scholar, we conclude that cookie respawning with browser fingerprinting lacks legal interpretation under the GDPR and the ePrivacy directive, but its use in practice may breach them, thus subjecting it to fines up to 20 million euro.

cs.CR↗

arXiv2020

Missed by Filter Lists: Detecting Unknown Third-Party Trackers with Invisible Pixels

Web tracking has been extensively studied over the last decade. To detect tracking, previous studies and user tools rely on filter lists. However, it has been shown that filter lists miss trackers. In this paper, we propose an alternative method to detect trackers inspired by analyzing behavior of invisible pixels. By crawling 84,658 webpages from 8,744 domains, we detect that third-party invisible pixels are widely deployed: they are present on more than 94.51% of domains and constitute 35.66% of all third-party images. We propose a fine-grained behavioral classification of tracking based on the analysis of invisible pixels. We use this classification to detect new categories of tracking and uncover new collaborations between domains on the full dataset of 4,216,454 third-party requests. We demonstrate that two popular methods to detect tracking, based on EasyList&EasyPrivacy and on Disconnect lists respectively miss 25.22% and 30.34% of the trackers that we detect. Moreover, we find that if we combine all three lists 379,245 requests originated from 8,744 domains still track users on 68.70% of websites.

cs.CR↗