The spread of fake news has had major impact on public discourse and trust. Detection methods rely heavily on evidence quality and verdict accuracy. Traditional approaches, often based on static sources, struggle with outdated or incomplete information, especially for new or obscure claims. Large Language Models (LLMs) offer promising reasoning and generation capabilities but face similar challenges, including outdated knowledge and limited coverage. To address these challenges, we present ClaimVerAgents1, a novel, retrieval-augmented, modular, and interpretable multi-agent system that leverages LLMs for real-time fake news verification. Each autonomous agent fulfills a specialized sub-task: claim extraction, query generation, evidence evaluation, verdict decision, and explanation generation, all within a transparent, confidenceaware pipeline. Extensive experiments on the PolitiFact dataset show that ClaimVerAgents outperforms both classical and LLM-based baselines in accuracy and robustness. Importantly, the system generates structured, humanreadable explanations alongside its verdicts, enhancing trust and interpretability.1Code and data are available at https://anonymous.4open.science/r/ ClaimVerAgents-832E

In today's digital era, the rapid spread of fake news undermines both social unity and democratic institutions, demanding effective countermeasures. Current browser extensions to counter fake news have significant limitations, such as opaque models, dependency on traditional Machine Learning (ML) techniques, lack of explanatory features, and limited focus on detection without user engagement support. This paper introduces Aletheia, a novel browser extension that addresses these shortcomings by leveraging Retrieval Augmented Generation (RAG) and Large Language Models (LLMs) to enhance fake news detection and provide evidence-based explanations. Additionally, Aletheia incorporates two key components: a Discussion Hub, enabling users to discuss instances of fake news, and a Stay Informed feature, which displays the latest fact-checks. Aletheia surpasses state-of-the-art methods according to experimental results.

The rise of online image sharing raises significant privacy concerns, as users may inadvertently disclose sensitive personal information. This paper introduces two Artificial Intelligence-driven systems, DEDALUS and ICARUS, designed to enhance user awareness and information security. DEDALUS is a modular image privacy classification system with advanced explainability capabilities. Its classification module, based on a voting ensemble of vision models, achieves good performance on an extended version of the PrivacyAlert dataset. Its explainability module combines a Large Language Model (LLM)-based risk assessment with LIME to interpret visual model decisions. To illustrate how DEDALUS’s capabilities can be integrated into a broader privacy-preserving application, we also introduce ICARUS, a system specialized in detecting privacy-related disclosures. It uses a vision model trained on a custom dataset to identify images containing Personally Identifiable Information (PII), and integrates the DEDALUS classifier via API chaining to issue user-friendly warnings. Two proof-of-concept applications are proposed: DEDALUS Workbench, for model analysis and dataset creation, and ICARUS Watcher, which simulates image sharing in unsecured environments. The ICARUS API, used in the Watcher demo, monitors image-sharing activity and issues alerts for both PII detection and private image classification. Together, these systems contribute to computer vision, AI ethics, and digital privacy by offering a novel approach to mitigating the risks associated with sharing sensitive visual content.

The Internet of Things (IoT) is an impending area with applications in numerous fields. The number of IoT devices has seen exponential growth, increasing apprehensions around security. Cyberattacks are of rising concern because of the expanded attack surface of threats that have plagued networks. Adding to that are insecure practices among users who may not know to protect their IoT devices. Therefore, IoT security has become fundamental, especially as IoT devices carry sensitive data. This paper provides a proof of concept of an Intelligent Intrusion Detection System for IoT. We centered our work on a privacy-preserving approach offering a Federated Learning (FL) based solution for intrusions recognition combining network and energy data. Our model has achieved high accuracy while preserving a short running time in multiple FL rounds.

Information and Communications Technology (ICT) is proliferating exponentially and has undoubtedly become an intrinsic part of our daily lives. However, its fast-paced growth has brought upon multiple challenges amongst which are the human-centric threats to cybersecurity and privacy. One of the main reasons for this is the shift in individuals’ behaviour towards carelessly disclosing private information, especially on social media.This work builds on the existing literature that identifies the motivations leading to oversharing in order to predict and mitigate self-disclosure. This paper aims to tackle this first by proposing a user model for the individual’s disclosure motivations. The aim is to measure how driven the user is to share personal information given a specific context. This is paramount to second objective, which is designing personalized privacy-preserving interventions known as nudges based on the user model. A visual aid is provided to further attract the user’s attention and persuade them to alter their behaviour. Study participants (N=800) were recruited via Mechanical Turk and responded to realistic scenarios to assess their motivations for sharing personal data. Then, persuasive nudges were pushed in the context of the evaluation.

The proliferation of social media set the foundation for the culture of over-disclosure where many people document every single event, incident, trip, etc. for everyone to see. Raising the individual's awareness of the privacy issues that they are subjecting themselves to can be challenging. This becomes more complex when the post being shared includes data "owned" by others. The existing approaches aiming to assist users in multi-party disclosure situations need to be revised to go beyond preferences to the "good" of the collective.

In this paper, we propose a new framework for mitigating biases in machine learning systems. The problem of the existing mitigation approaches is that they are model-oriented in the sense that they focus on tuning the training algorithms to produce fair results, while overlooking the fact that the training data can itself be the main reason for biased outcomes. Technically speaking, two essential limitations can be found in such model-based approaches: 1) the mitigation cannot be achieved without degrading the accuracy of the machine learning models, and 2) when the data used for training are largely biased, the training time automatically increases so as to find suitable learning parameters that help produce fair results. To address these shortcomings, we propose in this work a new framework that can largely mitigate the biases and discriminations in machine learning systems while at the same time enhancing the prediction accuracy of these systems. The proposed framework is based on conditional Generative Adversarial Networks (cGANs), which are used to generate new synthetic fair data with selective properties from the original data. We also propose a framework for analyzing data biases, which is important for understanding the amount and type of data that need to be synthetically sampled and labeled for each population group. Experimental results show that the proposed solution can efficiently mitigate different types of biases, while at the same time enhancing the prediction accuracy of the underlying machine learning model.

The present study explores the relationship between cybersecurity knowledge, online behavior, and risk perception. To simulate an online shopping experience, we invited participants to access our newly designed website and answer a set of questions to evaluate their level of cybersecurity knowledge. The experiment identifies multiple subsequent stages that enhance different features of the social online shopping literature by offering different tasks. The online activities provided on our website play the role of essential vital enablers to interact with the user such as giving the buyer the chance to earn more credits in exchange for private information, saving the virtual credit card information and selecting a specific shopping scenario. Our outcomes highlight the significance of engaging individuals with cybersecurity and provide an analysis of consumer profiling and practice in an online shopping context. Moreover, our findings lead to two sets of actions: reducing the perceived risk of cybercrime in online activities by increasing the level of knowledge in cybersecurity and evolving user behavior when spotting a high level of privacy concern.

We live in an age in which the dependency on technological tools is inescapable. At the same time, privacy-related issues are emerging in a way that we are at the breakpoint of losing control over our data. Information sharing by social-network, users can result in violations of privacy and security. For example, when a user is asking for a personalized service, he may find his contact details revealed, and may become the subject of harassment (cyber-bullying) or a potential victim of online deception or identity theft. Moreover, as Tim Berners-Lee stated, "The major players are making profit from our data. Therefore we lose out on the benefits we could realise if we had direct control over this data and choose when and with whom to share it". Today, more than ever, users need to keep control over their personal data when they ask for a personalized service. This tutorial, addresses how to reach this delicate balance between privacy and personalization.

This paper proposes an enhanced genetic algorithm (E-GA) with a new crossover operator for the well-known NP-hard traveling tournament problem (TTP). TTP is the problem of scheduling a feasible double round robin tournament that minimizes the total distances traveled by the teams. The proposed E-GA for TTP uses a new crossover operator based on sharing the best partial path teams among the schedules. Further, E-GA uses a variable neighborhood search as a subroutine to improve the intensification mechanism in GA. The proposed method is evaluated on publicly available standard benchmarks and compared with other techniques for TTP. The computational experiment shows that the proposed method could build interesting results comparable to other state-of-the-art approaches.

Nowadays, geosocial networks (GSNs) have become a significant component of people's daily lives as they are one of the most popular applications that are being widely accessed through smart devices such as smartphones and tablets. Their rapid widespread use and their invasion of our private life warrant a better understanding. In particular, the impact of trust in GSN, the privacy concerns of users, their perception of risk and the social influence on the use of such mobile applications is not yet fully understood. In this paper, we study the factors influencing the usage intention of GSN users. To realize this, we propose a model based on the user's perspective. Our model focuses on four overall factors that influence the users' concerns and in turn their intention and aim of using GSNs: privacy concerns, trust, social influence and risk perception. We tested empirically the proposed research model by running a web-based survey. The participants consisted of 396 persons with at least a past experience with GSNs. The results revealed that among all the possible factors the privacy concerns, social influence and trust have a significant impact on the intention and usage of GSNs. In contrast, personality traits have almost no effects on trust or social influence. One notable exception is computer self-efficacy that was found to induce a strong influence on the four principal factors.

In today's age of exposure, websites and Internet services are collecting personal data-with or without the knowledge or consent of users. Not only does new technology provide an abundance of methods for organizations to gather and store information, but people are also willingly sharing data with increasing frequency, exposing their intimate lives on social media websites such as Facebook, Twitter, You tube, My space and others. Moreover, online data brokers, search engines, data aggregators and many other actors of the web are profiling people for various purposes such as the improvement of marketing through better statistics and an ability to predict consumer behaviour. Other less known reasons include understanding the newest trends in education, gathering people's medical history or observing tendencies in political opinions. People who care about privacy use the Privacy Enhancing Technologies (PETs) to protect their data, even though clearly not sufficiently. Indeed, as soon as information is recorded in a database, it becomes permanently available for analysis. Consequently even the most privacy aware users are not safe from the threat of re-identification. On the other hand, there are many people who are willing to share their personal information, even when fully conscious of the consequences. A claim from the advocates of open access information is that the preservation of privacy should not be an issue, as people seem to be confortable in a world where their tastes, lifestyle or personality are digitized and publicly available. This paper deals with Internet data collection and voluntary information disclosure, with an emphasis on the problems and challenges facing privacy nowadays.

We show how the quantum paradigm can be used to speed up unsupervised learning algorithms. More precisely, we explain how it is possible to accelerate learning algorithms by quantizing some of their subroutines. Quantization refers to the process that partially or totally converts a classical algorithm to its quantum counterpart in order to improve performance. In particular, we give quantized versions of clustering via minimum spanning tree, divisive clustering and k -medians that are faster than their classical analogues. We also describe a distributed version of k -medians that allows the participants to save on the global communication cost of the protocol compared to the classical version. Finally, we design quantum algorithms for the construction of a neighbourhood graph, outlier detection as well as smart initialization of the cluster centres.

The number of phishing attacks against web serviceshas seen a steady increase causing, for example, a negative effecton the ability of banking and financial institutions to deliverreliable services on the Internet. This paper presents an automaticapproach detecting phishing attacks. Our approach combinesa personalized whitelisting approach with machine learningtechniques. The whitelist is used as filter that blocks phish webpages used to imitate innocuous user behavior. The phishingpages that are not blocked by the whitelist pass are furtherfiltered using a Support Vector Machine classifier designed andoptimized to classify these threats. Our experimental results showthat the proposed approach improves over the current state-ofthe-art methods.

Developers may introduce anti-patterns in their software systems because of time pressure, lack of understanding, communication, and--or skills. Anti-patterns impede development and maintenance activities by making the source code more difficult to understand. Detecting anti-patterns in a whole software system may be infeasible because of the required parsing time and of the subsequent needed manual validation. Detecting anti-patterns on subsets of a system could reduce costs, effort, and resources. Researchers have proposed approaches to detect occurrences of anti-patterns but these approaches have currently some limitations: they require extensive knowledge of anti-patterns, they have limited precision and recall, and they cannot be applied on subsets of systems. To overcome these limitations, we introduce SVMDetect, a novel approach to detect anti-patterns, based on a machine learning technique---support vector machines. Indeed, through an empirical study involving three subject systems and four anti-patterns, we showed that the accuracy of SVMDetect is greater than of DETEX when detecting anti-patterns occurrences on a set of classes. Concerning, the whole system, SVMDetect is able to find more anti-patterns occurrences than DETEX.

It is not uncommon for business managers to use recent innovations in information and communications technology to monitor employees and job candidates. These methods not only rely on heavy surveillance during working hours of employees but can also be applied outside their professional environment, to impinge on their personal lives. Surveillance techniques encompass such traditional means like recording cameras to more recent methods including analyzing social networks pages, performing extensive web searches and dealing with online data brokers. While monitoring initiatives set up by employers can have benefits for companies, the threat to privacy they entail can deteriorate the mental and physical health of employees and have a negative impact on the quality of relationship between colleagues. Businesses have a social responsibility and need to ensure that their behavior does not infringe upon their employee's rights to privacy. In this non-technical paper, we discuss some online approaches adopted by companies regarding employee surveillance. We elaborate on various methods employed by managers to monitor their employees and gain as much information as possible on job candidates. Then, these techniques are further discussed from the standpoint of their moral and legal perspectives with regards to privacy rights.

Identity theft has become one of the fastest growing crimes. Most people are unaware of the amount of data they disclose over all the Internet services proposed by search engines, social networking sites, e-commerce web sites, free online tools, etc. They are also unaware that this data can be easily aggregated, data-mined and linked together, which may lead to a potential identity theft should it fall into the wrong hands. If one adds up all of his online searching, communicating, shopping, browsing, blogging, chatting, reading and news sharing, one would realize that one revealed a complete picture of oneself and perhaps some information about his relatives, friends, colleagues, employer, etc. The potential value of this data is considerable for criminals. This paper deals with identity theft and all the issues raised by this type of computer crime. More precisely, it illustrates the variety of information that hackers may want to sift through, the attacks that they may perform and the locations where they can find the information.

The need for intelligent HCI has been reinforced by the increasing numbers of human-centered applications in our daily life. However, in order to respond adequately, intelligent applications must first interpret users’ actions. Identifying the context in which users’ interactions occur is an important step toward automatic interpretation of behavior. In order to address a part of this context-sensing problem, we propose a generic and application-independent framework for activity recognition of users interacting with a computer interface. Our approach uses Layered Hidden Markov Models (LHMM) and is based on eye-gaze movements along with keyboard and mouse interactions. The main contribution of the proposed framework is the ability to relate users’ interactions to a task model in variant applications and for different monitoring purposes. Experimental results from two user studies show that our activity recognition technique is able to achieve good predictive accuracy with a relatively small amount of training data.

Social Networking Sites (SNS), such as Facebook and LinkedIn, have become the established place for keeping contact with old friends and meeting new acquaintances. As a result, a user leaves a big trail of personal information about him and his friends on the SNS, sometimes even without being aware of it. This information can lead to privacy drifts such as damaging his reputation and credibility, security risks (for instance identity theft) and profiling risks. In this paper, we first highlight some privacy issues raised by the growing development of SNS and identify clearly three privacy risks. While it may seem a priori that privacy and SNS are two antagonist concepts, we also identified some privacy criteria that SNS could fulfill in order to be more respectful of the privacy of their users. Finally, we introduce the concept of a Privacy-enhanced Social Networking Site (PSNS) and we describe Privacy Watch, our first implementation of a PSNS.

Social networking sites (SNS) have become very popular during the past few years, as they allow users to both express their individuality and meet people with similar interests. Nonetheless, there are also many potential threats to privacy associated with these SNS such as identity theft and disclosure of sensitive information. However, many users still are not aware of these threats and the privacy settings provided by SNS are not flexible enough to protect user data. In addition, users do not have any control over what others reveal about them. As such, we conduct a preliminary study which examines the privacy protection issues on social networking sites (SNS) such as MySpace, Facebook and LinkedIn. Based on this study, we identify three privacy problems in SNS and propose a privacy framework as a foundation to cope with these problems.

In recent years, there has been an increase in research on privacy preserving e-Learning. As authors argue the need for privacy, there were no studies performed to corroborate this need. However, it is established that the learner's emotion do affect his learning. Therefore, this paper investigates the impact of privacy on the learner's emotional state. Particularly, we perform an experiment where participants perform web-based tests, in a no privacy and privacy enforced environments. We report as well on our analysis and findings.

Since their introduction, SNS (Social Networking Sites) such as MySpace, Facebook and LinkedIn have attracted millions of users and have become established places for keeping contact with old acquaintances and meeting new ones. Nonetheless, due to lack of user awareness and proper privacy protection tools, huge quantities of user data, including personal information, pictures and videos are quickly falling into the hands of authorities, strangers, recruiters and even the public at large. By using SNSs and accepting their privacy policy, users have volunteered to relinquish their ownership on their own data, which explains why the proposed privacy solutions based on current SNSs cannot solve all user privacy issues. As such, we start by setting the foundations for privacy and introduce a Privacy Framework for SNSs. Then, based on this framework, we present a User Privacy Policy (UPP) which provides users with an easy and flexible way to specify and communicate their privacy concerns to other users, third parties and to the SNS provider.

Recommender systems enable merchants to assist customers in finding products that best satisfy their needs. Unfortunately, current recommender systems suffer from various privacy-protection vulnerabilities. We report on the first experimental realization of a theoretical framework called ALAMBIC, which we had previously put forth to protect the privacy of customers and the commercial interests of merchants. Our system is a hybrid recommender that combines content-based, demographic and collaborative filtering techniques. The originality of our approach is to split customer data between the merchant and a semi-trusted third party, so that neither can derive sensitive information from their share alone. Therefore, the system can only be subverted by a coalition between these two parties. Experimental results confirm that the performance and user-friendliness of the application need not suffer from the adoption of such privacy-protection solutions. Furthermore, user testing of our prototype show that users react positively to the privacy model proposed.

The rapid evolution of our world means that learning and knowledge sharing are fast becoming a key challenge for individuals and organizations. In this paper, we present a system called HELP, whose aim is to locate information and recommend experts in organizations. Each user is being viewed simultaneously as an expert and a learner. We use two approaches: The first one consists of making the system retrieve one or several requests similar to the seeking-learner's request using a textual case-based reasoning technique. The second approach aims at locating experts in specific areas in order to recommend them to the users who request this expertise. For this purpose, we use a hybrid recommendation technique based on Collaborative Filtering (CF) and Case-Based Reasoning (CBR). In contrast to existing approaches in expertise location, we believe that CBR combined to CF enables HELP to better recommend expertise, taking into account the user's feedback concerning the technical and pedagogical skills of the experts.

By the term "quantization", we refer to the process of using quantum mechanics in order to improve a classical algorithm, usually by making it go faster. In this paper, we initiate the idea of quantizing clustering algorithms by using variations on a celebrated quantum algorithm due to Grover. After having introduced this novel approach to unsupervised learning, we illustrate it with a quantized version of three standard algorithms: divisive clustering, k-medians and an algorithm for the construction of a neighbourhood graph. We obtain a significant speedup compared to the classical approach.

Although E-learning has advanced considerably in the last decade, some of its aspects, such as E-testing, are still in the development phase. Authoring tools and test banks for E-tests are becoming an integral and indispensable part of E-learning platforms and with the implementation of E-learning standards, such as IMS QTI, E-testing material can be easily shared and reused across various platforms. With the knowledge available for re-use and exam automation comes a new challenge: making sure that created exams are free of conflicts. A Conflict exists in an exam if at least two questions within that exam are redundant in content, and/or if at least one question reveals the answer to another question within the same exam. In this paper we propose using Information Retrieval techniques to detect conflicts within an exam. Our solution, ICE (Identification of Conflicts in Exams), is based on the vector space model relying on tfidf weighing and the cosine function to calculate similarity. ICE also combines the hybrid recommendation techniques of the EQRS (Exam Question Recommender System) in order to propose replacements for conflicting questions.

We start with the usual paradigm in electronic commerce: a customer, Bob, wants to buy from a merchant, Alice. However, Bob wishes to enjoy maximal privacy while Alice needs to protect her sensitive data. Bob should be able to remain anonymous throughout the entire process, from turning on his comp uter to final delivery and even after-sale maintenance services. Ideally, he should even be able to hide from Alice what he is interested in buying. Conversely, Alice should not have to reveal anything unnecessary about her catalogue – especially prices – for fear that she might in fact be dealing with a hostile competitor masquerading as a customer. For this purpose, we introduce the Blind Electronic Commerce paradigm to offer an integrated solution to the dual conundrum of ensuring Bob's privacy as well as protecting Alice's sensitive information.

We start with the usual paradigm in electronic commerce: a consumer who wants to buy from a merchant. However, both parties wish to enjoy maximal privacy. In addition to remaining anonymous, the consumer wants to hide her browsing pattern and even the identification of the product she may decide to buy. Nevertheless, she wants to be able to negotiate the price, pay, receive the product and even enjoy maintenance on it. On the other hand, the merchant wants to leak as little information as possible on his catalogue for fear that he might in fact be dealing with a hostile competitor. For this purpose, we introduce the Blind Customer Buying Behaviour model, which adds confidentiality to the standard Customer Buying Behaviour model. In this paper, we concentrate on blind catalogue browsing.

The proper combination of hypermedia and interactive multimedia systems for a particular learning style can greatly enhance the learning environment. By interacting with this virtual environment, the learner can gain practical experience. We describe a Web-based training application based on adaptive hypermedia and interactive multimedia to train the users to configure and use a new software (SoftPhone). We propose an approach based on a learning style theory focused on "learning by doing" through active experimentation and reflective observation.