Cybergrooming is a form of online abuse that threatens teens’ mental health and physical safety. Yet, most prior work has focused on detecting perpetrators’ behaviors, leaving a limited understanding of how teens might respond to such unwanted advances. To address this gap, we conducted an online survey with 74 participants—51 parents and 23 teens—who responded to simulated cybergrooming scenarios in two ways: responses that they think would make teens more vulnerable or resilient to unwanted sexual advances. Through a mixed-methods analysis, we identified four types of vulnerable responses (encouraging escalation, accepting an advance, displaying vulnerability, and negating risk concern) and four types of protective strategies (setting boundaries, directly declining, signaling risk awareness, and leveraging avoidance techniques). As the cybergrooming risk escalated, both vulnerable responses and protective strategies showed a corresponding progression. This study contributes a teen-centered understanding of cybergrooming, a labeled dataset, and a stage-based taxonomy of perceived protective strategies, while offering implications for educational programs and sociotechnical interventions.

Cybergrooming is an evolving threat to youth, requiring proactive educational interventions. We address this by modeling dialogue progression as a structured planning problem over stage-wise interactions. We propose StagePilot, a dialogue framework that separates stage-level planning from response generation, in which the model selects the next stage under constrained transitions and generates responses conditioned on it, enabling coherent and realistic progression. Reinforcement learning is used to learn stage-level policies from offline data, optimizing for both emotional alignment and goal-consistent progression. Our empirical experiments show that StagePilot generates more structured, coherent dialogue trajectories and reduces conversational stagnation compared to baselines; notably, the IQL+AWAC variant reaches the final stage more often while maintaining over 70% positive or neutral responses, yielding a 43% relative improvement.

The robustness, security, and safety of artificial intelligence (AI) systems have become growing concerns, particularly as deep learning models are increasingly deployed in critical applications. Among emerging threats, backdoor attacks pose a serious risk by embedding hidden malicious behaviors into otherwise well-performing models. Although recent advances in detection techniques have improved defenses for computer vision systems, our findings demonstrate that even simple but carefully designed poisoning strategies can successfully evade these defenses. In this paper, we introduce GhostBackdoor, a novel backdoored model trained using a custom loss function and targeted data augmentation. The proposed loss function aligns neuron activations between clean and poisoned inputs, effectively masking activation anomalies, while the augmentation enforces strict location- and pattern-specific triggers that activate the backdoor only under specific conditions. After training, the model maintains behavior indistinguishable from a clean model unless exposed to the designated trigger with the specific pattern and at the designed locations. We evaluate GhostBackdoor against a broad range of leading defense mechanisms, most of which fail to detect the implanted backdoor. Our results highlight how the vast hypothesis space of deep learning models can be exploited to conceal malicious activations, underscoring the need for more robust security strategies in AI-driven systems, including those used in Internet of Things (IoT) applications.

Mission Impact Assessment (MIA) is critical for enhancing system effectiveness and ensuring mission success. This article presents Interdependent Mission Impact Assessment ( iMIA ), an interdependent MIA framework that models relationships among mission components and enables probabilistic reasoning under uncertainty. Designed for AI-driven mission systems operating in dynamic, low-data, or poorly observable environments, iMIA addresses the limitations of traditional methods that often rely on overly confident assumptions about adversary behavior. While conventional Hypergame Theory (HGT) captures perceptual uncertainty from asymmetric or inaccurate views, it overlooks epistemic uncertainty arising from limited knowledge. To bridge this gap, we introduce a hybrid Subjective Logic (SL)-based HGT model (SLHG), integrating SL to represent epistemic uncertainty and HGT to account for misperceptions. This integration supports informed decision-making under both uncertain strategy beliefs and divergent environmental views. iMIA evaluates mission impact using multidimensional system quality metrics, security, trust, resilience, and agility, across diverse attacker–defender interactions. It identifies critical nodes influencing mission outcomes and quantifies performance gains from asset capacity reinforcement and asset vulnerability mitigation. Applied to a vehicle-assisted AI-based mission system, iMIA with SLHG improves performance by 16% in \(ASR\) , 20% in \(MTBF\) , 11% in \(TSA\) , and 14% in \(P_{ACC}\) . Designed for incremental development, iMIA supports continuous feedback and iterative refinement. Our results show that feedback-driven adjustments improve overall system performance by up to 18% in the accuracy performance.

Federated Learning (FL) offers a promising approach for collaborative model training in healthcare while preserving data privacy. However, existing FL methods often fall short in addressing two critical challenges: client-level fairness and compounded uncertainty from data heterogeneity and privacy-preserving mechanisms. We propose fair-LDP, a fairness-aware Local Differential Privacy framework that promotes fairness and privacy via uncertainty-guided aggregation in federated healthcare AI. fair-LDP leverages evidential neural networks (ENNs) to quantify predictive uncertainty and introduces a novel strategy that uses uncertainty-driven local differential privacy to guide fairness-aware updates while preserving data privacy. This ensures equitable performance across clients with varying data quality while mitigating the influence of unreliable or outlier updates. fair-LDP incorporates an adaptive mechanism that adjusts each client's privacy budget based on model performance, balancing fairness, privacy, and accuracy. We evaluate fair-LDP on real-world healthcare datasets under both IID and non-IID settings. Our experimental results show that it consistently outperforms state-of-the-art fairness-aware and privacy-preserving FL baselines, with no added computational overhead, while maintaining privacy guarantees comparable to homomorphic encryption and secure multiparty computation. By integrating uncertainty modeling, fairness-aware aggregation, and adaptive local differential privacy, fair-LDP provides a practical and principled solution for responsible, equitable, and privacy-preserving federated learning in healthcare.

Uncertainty reasoning and quantification play a critical role in decision making across various domains, prompting increased attention from both academia and industry. As real-world applications become more complex and data-driven, effectively handling uncertainty becomes paramount for accurate and reliable decision making. This workshop focuses on the critical topics of uncertainty reasoning and quantification in decision making. It provides a platform for experts and researchers from diverse backgrounds to exchange ideas on cutting-edge techniques and challenges in this field. The interdisciplinary nature of uncertainty reasoning and quantification, spanning artificial intelligence, machine learning, statistics, risk analysis, and decision science, will be explored. The workshop aims to address the need for robust and interpretable methods for modeling and quantifying uncertainty, fostering reasoning decision-making in various domains. Participants will have the opportunity to share research findings and practical experiences, promoting collaboration and advancing decision-making practices under uncertainty.

Autonomous Vehicles (AVs) integrate numerous control units , network components, and protocols to operate effectively and interact with their surroundings, such as pedestrians and other vehicles. While these technologies enhance vehicle capabilities and enrich the driving experience, they also introduce new attack surfaces, making AVs vulnerable to cyber-attacks. Such cyber-attacks can lead to severe consequences, including traffic disruption and even threats to human life. Security modelling is crucial to safeguarding AVs as it enables the simulation and analysis of an AV’s security before any potential attacks. However, the existing research on AV security modelling methods for analysing security risks and evaluating the effectiveness of security measures remains limited. In this work, we introduce a novel graphical security model and metrics to assess the security of AV systems. The proposed model utilizes initial network information to build attack graphs and attack trees at different layers of network depth. From this, various metrics are automatically calculated to analyse the security and safety of the AV network. The proposed model is designed to identify potential attack paths, analyse security and safety with precise metrics, and evaluate various defence strategies. We demonstrate the effectiveness of our framework by applying it to two AV networks and distinct AV attack scenarios, showcasing its capability to enhance the security of AVs.

An accurate assessment of a mission system’s performance, called mission impact assessment (MIA), can effectively identify and counteract any issues in the current system to avoid potential risks or vulnerabilities that may cause mission failure. Although the importance of MIA research has been recognized to mitigate the system’s potential risk for more than a decade, little research has shown a comprehensive MIA framework with experimental validation showing the inference performance of the MIA tools under realistic attack–defense interactions. To fill this gap, we propose an interdependent MIA, callediMIA, that can adequately capture the interrelationships of the key components in a mission system and environment and accurately infer a mission outcome (i.e., success or failure) under uncertainty. In the proposediMIA, we first consider Subjective Bayesian Networks to evaluate how successfully a given system introduces impacts on mission success under uncertainty due to a lack of evidence. In addition, we consider strategic attack-defense interactions based on a hypergame theory in which the attacker and defender can take actions under their perceived uncertainty. Our extensive experiments showed the outperformance of our proposediMIAup to 70% and 25% over the performance of the baseline and the state-of-the-art Bayesian network (BN)-based counterparts, respectively, in terms of inference accuracy in mission performance (e.g., mission outcomes, such as mission success or failure). We also provide in-depth sensitivity analyses to identify the key system components that should be considered more carefully to avoid mission failure.

We propose a sustainable federated learning (FL)-based monitoring system, namely susFL, for smart animal farms to address the challenge of inconsistent health monitoring due to fluctuating energy levels of solar sensors. This system equips animals, such as cattle, with solar sensors with computational capabilities, including Raspberry Pis, to train a local deep-learning model on health data. These sensors periodically update Long Range (LoRa) gateways, forming a wireless sensor network (WSN) to detect diseases like mastitis. Our proposed susFL system incorporates a game-theoretic approach, called mechanism design, to select intelligent clients to optimize monitoring quality while minimizing energy use. This strategy ensures the system’s sustainability and resilience against various adversarial attacks, including data poisoning and privacy threats, that could disrupt FL operations. Our work in smart farm technologies sets a new standard by developing an animal monitoring system that is both energy-adaptive and resistant to attacks. Through extensive experiments, we demonstrate that our FL-based monitoring system significantly outperforms existing methods in prediction accuracy, operational efficiency, system reliability (i.e., mean time between failures or MTBF), and social welfare maximization by the mechanism designer. Our experimental results show that susFL significantly outperforms the state-of-the-art counterparts, including a 10% reduction in energy consumption, a 15% increase in social welfare, and a 34% rise in Mean Time Between Failures (MTBF) while maintaining the global model’s prediction accuracy.

Can we trust Large Language Models (LLMs) to accurately predict scam? This paper investigates the vulnerabilities of LLMs when facing adversarial scam messages for the task of scam detection. We addressed this issue by creating a comprehensive dataset with fine-grained labels of scam messages, including both original and adversarial scam messages. The dataset extended traditional binary classes for the scam detection task into more nuanced scam types. Our analysis showed how adversarial examples took advantage of vulnerabilities of a LLM, leading to high misclassification rate. We evaluated the performance of LLMs on these adversarial scam messages and proposed strategies to improve their robustness.

The Competitive Influence Maximization (CIM) problem involves entities competing to maximize influence in online social networks (OSNs). While Deep Reinforcement Learning (DRL) methods have shown promise, most assume binary user opinions and overlook behavioral factors. We introduce DRIM, a novel DRL-based CIM framework using Subjective Logic (SL) to incorporate user preferences and uncertainty, optimizing seed selection to spread true information while countering false information. DRIM’s Uncertainty-based Opinion Model (UOM) provides a realistic representation of user opinions. Results demonstrate that UOM maintains over 80% true influence against advanced misinformation, and DRIM outperforms state-of-the-art methods by up to 45% in influence and 77% in speed. DRIM also excels in limited-resource scenarios, networks with 10% invisibility, and when users are inclined to doubt true information.

Subjective Bayesian networks (SBN) integrate Bayesian Networks (BN) with Subjective Logic, enabling the representation of second-order uncertainty, denoting the uncertainty surrounding the probability distribution of an event. Although prior research predominantly centers on exact inference within the SBN framework, there is a notable dearth of exploration into the realm of approximate inference in SBN. Our work is specifically geared towards addressing this gap, focusing on the application of diverse sampling methodologies (i.e., forward and Gibbs sampling) for approximate inference in SBN. The primary contribution of this work lies not only in the introduction of approximate inference in SBN but also in the formulation of an “anytime” SBN inference algorithm. This implies that a best inference estimate can be obtained at any given moment, given trade-offs in the precision. Moreover, the allocation of computational resources is a customizable and potentially optimizable process. Through a rigorous series of experiments, we empirically demonstrate that the number of iterations to convergence decreases as we provide more samples for both forward and Gibbs sampling. Furthermore, we discover the difference between approximate and exact inference in belief ($\delta_{\text {belief }}$) and uncertainty ($\delta_{\text {uncertainty }}$) mass of subjective opinion becomes more unpredictable as the error gets large in BN probability. Lastly, in our experiments, we demonstrate the number of BN samples has a greater impact on $\delta_{\text {belief }}$ than the number of SBN iterations. These findings indicate that the family of greedy algorithms (based on local graded changes - such as gradients) can be a promising approach for finding optimal allocations of computational resources in this framework. The software assets produced and used in this work will be made available as an open source Python library.

Deep neural networks (DNNs) have been shown to perform well on exclusive, multi-class classification tasks. However, when different classes have similar visual features, it becomes challenging for human annotators to differentiate them. When an image is ambiguous, such as a blurry one where an annotator can't distinguish between a husky and a wolf, it may be labeled with both classes: {husky, wolf}. This scenario necessitates the use of composite set labels. In this paper, we propose a novel framework called Hyper-Evidential Neural Network (HENN) that explicitly models predictive uncertainty caused by composite set labels in training data in the context of the belief theory called Subjective Logic (SL). By placing a Grouped Dirichlet distribution on the class probabilities, we treat predictions of a neural network as parameters of hyper-subjective opinions and learn the network that collects both single and composite evidence leading to these hyper-opinions by a deterministic DNN from data. We introduce a new uncertainty type called vagueness originally designed for hyper-opinions in SL to quantify composite classification uncertainty for DNNs. Our experiments prove that HENN outperforms its state-of-the-art counterparts based on four image datasets. The code and datasets are available at: https://shorturl.at/dhoqx.

Uncertainty reasoning and quantification play a critical role in decision making across various domains, prompting increased attention from both academia and industry. As real-world applications become more complex and data-driven, effectively handling uncertainty becomes paramount for accurate and reliable decision making. This workshop focuses on the critical topics of uncertainty reasoning and quantification in decision making. It provides a platform for experts and researchers from diverse backgrounds to exchange ideas on cutting-edge techniques and challenges in this field. The interdisciplinary nature of uncertainty reasoning and quantification, spanning artificial intelligence, machine learning, statistics, risk analysis, and decision science, will be explored. The workshop aims to address the need for robust and interpretable methods for modeling and quantifying uncertainty, fostering reasoning decision-making in various domains. Participants will have the opportunity to share research findings and practical experiences, promoting collaboration and advancing decision-making practices under uncertainty.

Smart farms, as a way for better productivity and efficiency, have yet to be thoroughly studied for their service quality amid cyber threats. This work introduces a proactive security approach, Moving Target Defense (MTD), to the smart farm system to proactively address diverse cyber threats. Specifically, we develop an energy-aware MTD, termed eMTD, using port hopping for a sustainable smart farm network. Leveraging deep reinforcement learning (DRL), we identify the optimal MTD strategy capable of ensuring high monitoring quality of animal conditions and sufficient energy levels for solar-powered sensors on the farm. Our experiments demonstrate a significant improvement by approximately 15% in monitoring quality and remaining energy compared to other schemes.

A mission impact assessment (MIA) framework assesses a mission system’s performance and/or aims to identify risk factors of mission failure to take a mitigation strategy. The iMIA framework, unlike traditional MIA approaches, comprehensively addresses the interdependencies of key system components like asset vulnerabilities, attack behaviors, defense mechanisms, and service/task characteristics. This framework goes beyond conceptual models, providing a validated and detailed MIA framework covering from a conceptual model to detailed mission and system designs. Existing MIA approaches with Bayesian Networks (BNs) overlook inherent uncertainties arising from factors like insufficient evidence or data in real-world applications. To address decision-making challenges in the face of uncertainties, we introduce Subjective Bayesian Networks (SBNs). SBNs estimate uncertainties and interpret them using an expert’s domain knowledge or historical data, forming a subjective opinion through prior belief in SBNs. Using the SBNs, we enhance iMIA’s inference accuracy in a Federated Learning-based mission system (FLMS) for vehicular networks. Our experiments show thatiMIA’s SBN reasoning significantly improves mission outcome inference accuracy compared to conventional BN-based reasoning in existing MIA approaches. We also evaluate mission performance based on service availability and prediction accuracy from the FLMS.

While smart farm technologies significantly aid in reducing costs and boosting productivity for farmers, they often lack the necessary robustness against cyberattacks and adaptability to dynamic environmental changes. We propose a solar-powered sensor-based smart farm system to provide high monitoring quality while preserving sensor energy in the presence of adversarial attacks. In a smart farm system, solar-powered sensors are attached to animals (e.g., cows) to monitor their health under varying weather conditions to provide energy-adaptive and high-quality monitoring services. Further, a smart farm system should be robust against adversarial attacks aiming to disrupt monitoring quality. We use deep reinforcement learning (DRL) to identify the optimal policy for maximizing monitoring quality and prolonging the system’s lifetime while maintaining sufficient energy. We introduce transfer learning (TL) into the DRL process to achieve fast learning without experiencing a cold start problem in DRL. In addition, we develop an uncertainty-aware anomaly data detection method to filter out deceptive data caused by adversarial attacks. Via extensive comparative performance analysis conducted based on real datasets, we demonstrate the superior performance of the proposed TL-based DRL strategies over existing competitive counterparts in the system lifetime, the monitoring quality, the learning convergence time, and the energy consumption.

As online social networks (OSNs) become more prevalent, a new paradigm for problem-solving through crowd-sourcing has emerged. By leveraging the OSN platforms, users can post a problem to be solved and then form a team to collaborate and solve the problem. A common concern in OSNs is how to form effective collaborative teams, as various tasks are completed through online collaborative networks. A team’s diversity in expertise has received high attention to producing high team performance in developing team formation (TF) algorithms. However, the effect of team diversity on performance under different types of tasks has not been extensively studied. Another important issue is how to balance the need to preserve individuals’ privacy with the need to maximize performance through active collaboration, as these two goals may conflict with each other. This research has not been actively studied in the literature. In this work, we develop a TF algorithm in the context of OSNs that can maximize team performance and preserve team members’ privacy under different types of tasks. Our proposed PRivAcy-Diversity-Aware TF framework, called PRADA-TF , is based on trust relationships between users in OSNs where trust is measured based on a user’s expertise and privacy preference levels. The PRADA-TF algorithm considers the team members’ domain expertise, privacy preferences, and the team’s expertise diversity in the process of TF. Our approach employs game-theoretic principles Mechanism Design to motivate self-interested individuals within a TF context, positioning the mechanism designer as the pivotal team leader responsible for assembling the team. We use two real-world datasets (i.e., Netscience and IMDb) to generate different semi-synthetic datasets for constructing trust networks using a belief model (i.e., Subjective Logic) and identifying trustworthy users as candidate team members. We evaluate the effectiveness of our proposed PRADA-TF scheme in four variants against three baseline methods in the literature. Our analysis focuses on three performance metrics for studying OSNs: social welfare, privacy loss, and team diversity.

This work aims to develop a surveillance mission system using unmanned aerial vehicles (UAVs) or drones when Denial-of-Service (DoS) attacks are present to disrupt normal operations for mission systems. In particular, we introduce the concept of cyber deception using honey drones (HDs) to protect the mission system from DoS attacks. HDs exhibit fake vulnerabilities and employ stronger signal strengths to lure DoS attacks, unlike the legitimate drones called mission drones (MDs) deployed for mission execution. This research formulates an optimization problem to identify an optimal set of signal strengths of HDs and MDs to best prevent the system from DoS attacks while maximizing mission performance under the resource constraints of UAVs. To solve this optimization problem, we leverage deep reinforcement learning (DRL) to achieve these multiple objectives of the mission system concerning system security and performance. Particularly, for efficient and effective parallel processing in DRL, we utilize a DRL algorithm called the Asynchronous Advantage Actor-Critic (A3C) algorithm to model attack-defense interactions. We employ a physical engine-based simulation testbed to consider realistic scenarios and demonstrate valid findings from the realistic testbed. The extensive experiments proved that our HD-based approach could achieve up to a 32% increase in mission completion, a 20% reduction in energy consumption, and a 62% decrease in attack success rates compared to existing defense strategies.

Subjective Bayesian networks (SBN) combine Bayesian Networks (BN) with Subjective Logic in order to express the second-order uncertainty (i.e., the uncertainty about a probability distribution of an event – as opposed to the uncertainty about the event itself). While SBNs provide a strong formalism for treating the uncertainty in a higher level, the literature lacks support for extensive software implementations focused on compatibility with current software solutions or standards. Our work explores the structural congruence between BN and SBN (in terms of software data structure) and a semantic bijection between Subjective Logic opinions to Dirichlet distributions to introduce a SBN reasoning framework that targets on effectively reusing the existing BN solutions. Particularly, we developed two inference algorithms that apply a Monte Carlo method to existing BN inference algorithms (we chose the Junction Tree algorithm, for test), respectively for batch and interactive SBN reasoning. A method for translating an evidence in SBN to uncertain (virtual) evidence in BN is also presented. The main contribution of this paper is the introduction of a simple, yet flexible empirical estimation method and a software architecture that virtually adapts any BN inference algorithm to an approximate computational inference framework for SBN. We also developed a Java component to demonstrate the reusability, and we developed a case study of Mission Impact Assessment of Unmanned Aerial Vehicles transporting critical items between hospitals in order to illustrate the applicability in a knowledge engineering process.

Early event detection aims to detect events even before the event is complete. However, most of the existing methods focus on an event with a single label but fail to be applied to cases with multiple labels. Another non-negligible issue for early event detection is a prediction with overconfidence due to the high vacuity uncertainty that exists in the early time series. It results in an over-confidence estimation and hence unreliable predictions. To this end, technically, we propose a novel framework, Multi-Label Temporal Evidential Neural Network (MTENN), for multi-label uncertainty estimation in temporal data. MTENN is able to quality predictive uncertainty due to the lack of evidence for multi-label classifications at each time stamp based on belief/evidence theory. In addition, we introduce a novel uncertainty estimation head (weighted binomial comultiplication (WBC)) to quantify the fused uncertainty of a sub-sequence for early event detection. We validate the performance of our approach with state-of-the-art techniques on real-world audio datasets.

Federated Learning (FL) has been adopted in practical network applications and plays a critical role. As FL allows participants to contribute to the global model by training locally with private data, it is known particularly vulnerable to neural backdoor attacks. This paper proposes a new defense, ScanFed, against neural backdoor attacks to FL systems. It leverages the synchronous nature of FL to effectively single out malicious neuron candidates and further validate if they indeed hijack the model's behaviors. Compared to existing neural backdoor defenses, ScanFed has the following distinct properties. First, it is extremely computation-friendly that is six orders of magnitude faster than state-of-the-art behavior-based backdoor defenses, rendering it highly suitable for large-scale FL systems. Second, it inherits the precise nature of behavior-based backdoor detection, making it significantly more effective than similarity-based defenses against advanced attacks. Third, it is robust to biased models uploaded by clients with non-IID (Independent and Identically Distributed) data, which is very common in practical FL systems. In addition, it is a plug-n-play scheme that can be seamlessly integrated into existing FL systems. To the best of our knowledge, this is the first behavior-based defense that enables scalable, efficient and accurate neural backdoor detection of FL systems in non-IID scenarios. This work delivers a ScanFed prototype and fully tests it in various settings of datasets, neural architectures, and backdoor attacks. The experiments demonstrate ScanFed achieves competitive accuracy and minimal detection time.

This paper deals with a cybergrooming and sexual misconduct topic in artificial intelligence-based educational programs. Although cybergrooming has been recognized as a cybercrime, there is a lack of programs to protect youth from cybergrooming. We present a generative chatbot framework, SERI (Stop cybERgroomIng), that can generate fluent authentic conversations in the context of cybergrooming between a perpetrator chatbot and a potential victim chatbot. Furthermore, we propose deep-reinforcement-learning-based dialogue generation with a stage-related reward to lead the conversation to the expected stage. We also minimize potential ethical issues introduced by the perverted languages when deploying the chatbots for cybersecurity education programs. We evaluated the conversations of SERI with open-source referenced, unreferenced metrics and human evaluation. We developed SERI as a platform for deploying perpetrator chatbot to interact with youth users to observe their responses and collect reactions when they are asked for private or sensitive information by the perpetrator.

Intent mining is critical for controlling the spread of false information across online social networks (OSNs). To this end, we develop deep reinforcement learning (DRL) agents guided by a delayed reward based on intent prediction using a classifier of long short-term memory (LSTM). Additionally, we incorporate an uncertainty-aware function that leverages subjective opinions derived from Subjective Logic (SL). Through evaluation using an annotated fake news tweet dataset, our results demonstrate that our intent classification framework surpasses competing methods in terms of intent accuracy. Our intent mining solutions using DRL algorithms can support effective and efficient intervention strategies for fake news spreading on OSNs.

Uncertainty reasoning and quantification play a critical role in decision making across various domains, prompting increased attention from both academia and industry. As real-world applications become more complex and data-driven, effectively handling uncertainty becomes paramount for accurate and reliable decision making. This workshop focuses on the critical topics of uncertainty reasoning and quantification in decision making. It provides a platform for experts and researchers from diverse backgrounds to exchange ideas on cutting-edge techniques and challenges in this field. The interdisciplinary nature of uncertainty reasoning and quantification, spanning artificial intelligence, machine learning, statistics, risk analysis, and decision science, will be explored. The workshop aims to address the need for robust and interpretable methods for modeling and quantifying uncertainty, fostering reasoning decision-making in various domains. Participants will have the opportunity to share research findings and practical experiences, promoting collaboration and advancing decision-making practices under uncertainty.

Mission Impact Assessment (MIA) is a critical endeavor for evaluating the performance of mission systems, encompassing intricate elements such as assets, services, tasks, vulnerability, attacks, and defenses. This study introduces an innovative MIA framework that transcends existing methodologies by intricately modeling the interdependencies among these components. Additionally, we integrate hypergame theory to address the strategic dynamics of attack-defense interactions. To illustrate its practicality, we apply the framework to an Internet-of-Things (IoT)-based mission system tailored for accurate, time-sensitive object detection. Rigorous simulation experiments affirm the framework's robustness across a spectrum of scenarios. Our results prove that the developed MIA framework shows a sufficiently high inference accuracy (e.g., 80 %) even with a small portion of the training dataset (e.g., 20–50 %).

Unmanned Aerial Vehicles (UAVs) have been used for surveillance operations, search and rescue missions, and delivery services. Given their importance and versatility, they naturally become targets for cyberattacks. Denial-of-Service (DoS) attacks are commonly considered to exhaust their resources or crash UAVs (or drones). This work proposes a unique proactive defense using honey drones (HD) for UAVs during surveillance operations. These HDs use lightweight virtual machines to lure and redirect potential DoS attacks. Both the choice of target by the attacker and the HD's deceptive tactics are influenced by the strength of the radio signal. However, a critical trade-off exists in that stronger signals can deplete battery life, while weaker signals can negatively affect the connectivity of a drone fleet network. To address this, we formulate an optimization problem to select the best strategies for an attacker or defender in selecting their signal strength level. We propose a novel HD-based defense to identify the optimal setting using deep reinforcement learning (DRL) or game theory and compare their performance with that of non-HD-based methods, such as Intrusion Detection Systems and ContainerDrone. Our experiments demonstrate the unique benefits and superior efficacy of each HD-based defense across various attack scenarios.

We propose end-to-end multimodal fact-checking and explanation generation, where the input is a claim and a large collection of web sources, including articles, images, videos, and tweets, and the goal is to assess the truthfulness of the claim by retrieving relevant evidence and predicting a truthfulness label (e.g., support, refute or not enough information), and to generate a statement to summarize and explain the reasoning and ruling process. To support this research, we construct MOCHEG, a large-scale dataset consisting of 15,601 claims where each claim is annotated with a truthfulness label and a ruling statement, and 33,880 textual paragraphs and 12,112 images in total as evidence. To establish baseline performances on MOCHEG, we experiment with several state-of-the-art neural architectures on the three pipelined subtasks: multimodal evidence retrieval, claim verification, and explanation generation, and demonstrate that the performance of the state-of-the-art end-to-end multimodal fact-checking does not provide satisfactory outcomes. To the best of our knowledge, we are the first to build the benchmark dataset and solutions for end-to-end multimodal fact-checking and explanation generation. The dataset, source code and model checkpoints are available at https://github.com/VT-NLP/Mocheg.

This work proposes an energy-adaptive monitoring system for a smart farm using solar sensors attached to cows. The proposed system aims to achieve a high monitoring quality in the smart farm under fluctuating energy and cyber attacks disrupting the collection of sensed data from solar sensors, such as protocol noncompliance, false data injection, denial-of-service, and state manipulation. We adopt Subjective Logic, a belief model, to consider multidimensional uncertainty in sensed data. We employ deep reinforcement learning (DRL) for agents on gateways to collect high-quality sensed data from the solar sensors. The DRL agents aim to collect high-quality sensed data with low uncertainty and high freshness under fluctuating energy levels in solar sensors. We analyze the performance of the proposed energy-adaptive smart farm system in accumulated reward, monitoring error rate, and system overload. We conduct a comparative performance analysis of the uncertainty-aware DRL algorithms against their counterparts in choosing the number of sensed data to be updated to collect high-quality sensed data to achieve high resilience against attacks. Our results prove that multiagent proximal policy optimization (MAPPO) using the uncertainty maximization technique outperforms other counterparts, showing about 4% lower monitoring error rate and the system overload.

Machine learning (ML)-based intrusion detection system (IDS) approaches have been significantly applied and advanced the state-of-the-art system security and defense mechanisms. In smart grid computing environments, security threats have been significantly increased as shared networks are commonly used, along with the associated vulnerabilities. However, compared to other network environments, ML-based IDS research in a smart grid is relatively unexplored, although the smart grid environment is facing serious security threats due to its unique environmental vulnerabilities. In this article, we conducted an extensive survey on ML-based IDS in smart grids based on the following key aspects: (1) The applications of the ML-based IDS in transmission and distribution side power components of a smart power grid by addressing its security vulnerabilities; (2) dataset generation process and its usage in applying ML-based IDSs in the smart grid; (3) a wide range of ML-based IDSs used by the surveyed papers in the smart grid environment; (4) metrics, complexity analysis, and evaluation testbeds of the IDSs applied in the smart grid; and (5) lessons learned, insights, and future research directions.

For decades, the spread of disinformation in online social networks (OSNs) has been a serious social issue. Disinformation via social media can easily mislead people’s beliefs toward or against an event that may mislead their behaviors based on the misbeliefs. The game theory approaches have been proposed under dynamic settings to limit the adverse influences of disinformation. It is a challenge to expand the users’ game strategies from the spreading decisions to the possible opinion updating choices. This work proposes a game-theoretic opinion framework that can formulate dynamic opinions by a belief model called Subjective Logic (SL) and provide opinion updates on five types of users’ interactions on OSN platforms. The opinions are updated based on user choices and user types through the game interactions among legitimate users, attackers, and a defender in an OSN. Via the extensive simulation experiments, the effectiveness of the opinion models of five decision-makers (DMs) is analyzed in terms of users believing or disbelieving disinformation in an epidemic model with parameter optimization. Our results show that while homophily-based DMs (H-DMs) introduce the highest opinion polarization, uncertainty-based DMs (U-DMs) can effectively filter untrustworthy users propagating disinformation.

Existing defensive deception (DD) approaches apply game theory, assuming that an attacker and defender play the same, full game with all possible strategies. However, in deceptive settings, players may have different beliefs about the game itself. Such structural uncertainty is not naturally handled in traditional game theory. In this work, we formulate an attackdefense hypergame where multiple advanced persistent threat (APT) attackers and a single defender play a repeated game with different perceptions. The hypergame model systematically evaluates how various DD strategies can defend proactively against APT attacks. We present an adaptive method to select an optimal defense strategy using hypergame theory for strategic defense as well as machine learning for adaptive defense. We conducted in-depth experiments to analyze the performance of the eight schemes including ours, baselines, and existing counterparts. We found the DD strategies showed their highest advantages when the hypergame and machine learning are considered in terms of reduced false positives and negatives of the NIDS, system lifetime, and players’ perceived uncertainties and utilities. We also analyze the Hyper Nash Equilibrium of given hypergames and discuss the key findings and insights behind them.

Deep learning-based cybersecurity applications should be able to continually accumulate threat knowledge for new types of threats over time while maintaining the knowledge of threats already exposed to the application. This paper proposes episodic memory management for continual learning with network intrusion datasets. For new attacks, the number of samples may not be sufficiently large for training, and thus the memory management algorithm should retain as many samples as possible instead of random sampling in the episodic memory for continual learning. The experiment results indicated that the proposed algorithm outperforms offline learning in terms of average per-class accuracy in a continual scenario with a network intrusion dataset.

Malicious reconnaissance is a critical step for attackers to collect sufficient network knowledge and choose valuable targets for intrusion. Defensive deception (DD) is an essential strategy against threats by misleading attackers' observations and beliefs. Honeypots are widely used for cyber deception that aims to confuse attackers and waste their resources and efforts. Defenders may use low-interaction honeypots or high-interaction honeypots. In this paper, we consider a hybrid honeypot system that balances the use of the two levels of honeypot complexity, where high-interaction honeypots are more capable of deceiving skilled attackers than low-interaction honeypots. We present a two-player hypergame model that characterizes how a defender should deploy low and high-interaction honeypots to defend the network against malicious reconnaissance activities. We model the tradeoff of each player and characterize their best strategies within a hypergame framework that considers the imperfect knowledge of each player toward their opponent. Finally, our numerical results validate the effectiveness of the proposed honeypot system.

In this work, we propose an energy-adaptive moni-toring system for a solar sensor-based smart animal farm (e.g., cattle). The proposed smart farm system aims to maintain high-quality monitoring services by solar sensors with limited and fluctuating energy against a full set of cyberattack behaviors including false data injection, message dropping, or protocol non-compliance. We leverage Subjective Logic (SL) as the belief model to consider different types of uncertainties in opinions about sensed data. We develop two Deep Reinforcement Learning (D RL) schemes leveraging the design concept of uncertainty maximization in SL for DRL agents running on gateways to collect high-quality sensed data with low uncertainty and high freshness. We assess the performance of the proposed energy-adaptive smart farm system in terms of accumulated reward, monitoring error, system overload, and battery maintenance level. We compare the performance of the two DRL schemes developed (i.e., multi-agent deep Q-Iearning, MADQN, and multi-agent proximal policy optimization, MAPPO) with greedy and random baseline schemes in choosing the set of sensed data to be updated to collect high-quality sensed data to achieve resilience against attacks. Our experiments demonstrate that MAPPO with the uncertainty maximization technique outperforms its counterparts.

As cyberattacks continuously threaten conventional defense techniques, Moving Target Defense (MTD) has emerged as a promising countermeasure to defend a system against them by dynamically changing attack surfaces of the system. MTD provides the system a state-of-art security mechanism that increases the attack cost or complexity of the system aiming for reducing vulnerabilities exposed to potential attackers. However, the notion of the proactive and dynamic systems adopting MTD services causes a substantial trade-off between system performance and security effectiveness, compared to conventional defense strategies. The MTD tactics accordingly result in performance degradation (e.g., interruptions of service availability) as one of the drawbacks caused by continuous mutations of the system configuration. Therefore, it is crucial to validate not only the security benefits against system threats but also quality-of-service (QoS) for clients when an MTD-enabled system proactively continues to mutate attack surfaces. This paper contributes to (i) developing new security metrics; (ii) measuring both the performance degradation and security effectiveness against potential real attacks (i.e., scanning, HTTP flood, dictionary, and SQL injection attack); and (iii) comparing the proposed job management strategies (i.e., drop and switch-over) from a performance and security perspective in a physical SDN testbed.

As cyberattacks are rising, Moving Target Defense (MTD) can be a countermeasure to proactively protect a networked system against cyber-attacks. Despite the fact that MTD systems demonstrate security effectiveness against the reconnaissance of Cyber Kill Chain (CKC), a time-based MTD has a limitation when it comes to protecting a system against the next phases of CKC. In this work, we propose a novel hybrid MTD technique, its implementation and evaluation. Our hybrid MTD system is designed on a real SDN testbed and it uses an intrusion detection system (IDS) to provide an additional MTD triggering condition. This in itself presents an extra layer of system protection. Our hybrid MTD technique can enhance security in the response to multi-phased cyber-attacks. The use of the reactive MTD triggering from intrusion detection alert shows that it is effective to thwart the further phase of detected cyber-attacks. We also investigate the performance degradation due to more frequent MTD triggers.This work contributes to (1) proposing an ML-based rule classification model for predicting identified attacks which helps a decision-making process for security enhancement; (2) developing a hybrid-based MTD integrated with a Network Intrusion Detection System (NIDS) with the consideration of performance and security; and (3) assessment of the performance degradation and security effectiveness against potential real attacks (i.e., scanning, dictionary, and SQL injection attack) in a physical testbed.

Fault diagnosis plays a critical role in maintaining and troubleshooting engineered systems. Various diagnosis models, such as Bayesian networks (BNs), have been proposed to deal with this kind of problem in the past. However, the diagnosis results may not be reliable if second-order uncertainty is involved. This article proposes a hierarchical system diagnosis fusion framework that considers the uncertainty based on a belief model, called subjective logic (SL), which explicitly deals with uncertainty representing a lack of evidence. The proposed system diagnosis fusion framework consists of three steps: 1) individual subjective BNs (SBNs) are designed to represent the knowledge architectures of individual experts; 2) experts are clustered as expert groups according to their similarity; and 3) after inferring expert opinions from respective SBNs, the one opinion fusion method was used to combine all opinions to reach a consensus based on the aggregated opinion for system diagnosis. Via extensive simulation experiments, we show that the proposed fusion framework, consisting of two operators, outperforms the state-of-the-art fusion operator counterparts and has stable performance under various scenarios. Our proposed fusion framework is promising for advancing state-of-the-art fault diagnosis of complex engineered systems.

Reinforcement learning (RL) is a promising approach for intelligent agents to protect a given system under highly hostile environments. RL allows the agent to adaptively make sequential defense decisions based on the perceived current state of system security aiming to achieve the maximum defense performance in terms of fast, efficient, and automated detection, threat analysis, and response to the threat. In this paper, we propose a deep reinforcement learning (DRL)-based adaptive traffic inspection and moving target defense countermeasure framework, called ‘DIVERGENCE,’ for building a secure networked system. The DIVERGENCE provides two main security services: (1) a DRL-based network traffic inspection mechanism to achieve scalable and intensive network traffic visibility for rapid threat detection; and (2) an address shuffling-based moving target defense (MTD) technique to defend against threats as a proactive intrusion prevention mechanism. Through extensive simulations and experiments, we demonstrate that the DIVERGENCE successfully caught malicious traffic flows while significantly reducing the vulnerability of the network through MTD.

Defensive deception techniques have emerged as a promising proactive defense mechanism to mislead an attacker and thereby achieve attack failure. However, most game-theoretic defensive deception approaches have assumed that players maintain consistent views under uncertainty. They do not consider players’ possible, subjective beliefs formed due to asymmetric information given to them. In this work, we formulate a hypergame between an attacker and a defender where they can interpret the same game differently and accordingly choose their best strategy based on their respective beliefs. This gives a chance for defensive deception strategies to manipulate an attacker’s belief, which is the key to the attacker’s decision-making. We consider advanced persistent threat (APT) attacks, which perform multiple attacks in the stages of the cyber kill chain (CKC) where both the attacker and the defender aim to select optimal strategies based on their beliefs. Through extensive simulation experiments, we demonstrated how effectively the defender can leverage defensive deception techniques while dealing with multi-staged APT attacks in a hypergame in which the imperfect information is reflected based on perceived uncertainty, cost, and expected utilities of both the attacker and defender, the system lifetime (i.e., mean time to security failure), and improved false-positive rates of intrusion detection.

Diversity-based security approaches have been studied for several decades since the 1970s. The concept ofdiversity-by-designemerged in the 1980s. Since then, diversity-based system design research has been explored to provide more secure and dependable services in cyber-physical systems (CPSs). In this work, we are particularly interested in providing an in-depth, comprehensive survey of existing diversity-based approaches, their insights, and associated future work directions for building secure and dependable CPSs. This will allow us to provide promising ways of providing quality network and services based on key diversity-by-design principles for those who want to conduct research on developing secure and dependable CPSs using diversity as a system design feature. This survey paper mainly provides: (i) The common concept of diversity based on its multidisciplinary nature along with the historical evolution of the concept of diversity-by-design for providing secure and dependable services; (ii) the key diversity-by-design principles; (iii) the key benefits and caveats of using the diversity-by-design; (iv) the main concerns of CPS environments utilizing the diversity-by-design; (v) an extensive survey and discussions of existing diversity-based approaches based on five different classifications; (vi) the types of attacks considered by diversity-based approaches; (vii) the overall trends of evaluation methodologies used for diversity-based approaches, in terms of metrics, datasets, and testbeds; and (viii) the insights, lessons, and gaps identified from this extensive survey and future work directions.

Resource constrained Internet-of-Things (IoT) devices are highly likely to be compromised by attackers, because strong security protections may not be suitable to be deployed. This requires an alternative approach to protect vulnerable components in IoT networks. In this article, we propose an integrated defense technique to achieve intrusion prevention by leveraging cyberdeception (i.e., a decoy system) and moving target defense (i.e., network topology shuffling). We evaluate the effectiveness and efficiency of our proposed technique analytically based on a graphical security model in a software-defined networking (SDN)-based IoT network. We develop four strategies (i.e., fixed/random and adaptive/hybrid) to address “when” to perform network topology shuffling and three strategies (i.e., genetic algorithm/decoy attack path-based optimization/random) to address “how” to perform network topology shuffling on a decoy-populated IoT network, and we analyze which strategy can best achieve a system goal, such as prolonging the system lifetime, maximizing deception effectiveness, maximizing service availability, or minimizing defense cost. We demonstrated that a software-defined IoT network running our intrusion prevention technique at the optimal parameter setting prolongs system lifetime, increases attack complexity of compromising critical nodes, and maintains superior service availability compared with a counterpart IoT network without running our intrusion prevention technique. Further, when given a single goal or a multi-objective goal (e.g., maximizing the system lifetime and service availability while minimizing the defense cost) as input, the best combination of “when” and “how” strategies is identified for executing our proposed technique under which the specified goal can be best achieved.

The growing number of smart vehicles makes it possible to envision a crowdsensing service where vehicles can share video data of their surroundings for seeking out traffic conditions and car accidents ahead. However, the service may need to deal with situations like malicious vehicles propagating false information to divert other vehicles to arrive at destinations earlier or lead them to dangerous locations. This article proposes a context-aware trust estimation scheme that can allow roadside units in a vehicular edge network to provide real-time crowdsensing services in a reliable manner by selectively using information from trustworthy sources. Our proposed scheme is novel in that its trust estimation does not require any prior knowledge of vehicles on roads but quickly obtains the accurate trust value of each vehicle by leveraging transfer learning. and its Q-learning-based dynamic adjustment scheme autonomously estimates trust levels of oncoming vehicles with the aim of detecting malicious vehicles and accordingly filtering out untrustworthy input from them. Based on an extensive simulation study, we prove that the proposed scheme outperforms existing ones in terms of malicious vehicle detection accuracy.

Traditional deep neural networks (NNs) have significantly contributed to the state-of-the-art performance in the task of classification under various application domains. However, NNs have not considered inherent uncertainty in data associated with the class probabilities where misclassification under uncertainty may easily introduce high risk in decision making in real-world contexts (e.g., misclassification of objects in roads leads to serious accidents). Unlike Bayesian NN that indirectly infer uncertainty through weight uncertainties, evidential NNs (ENNs) have been recently proposed to explicitly model the uncertainty of class probabilities and use them for classification tasks. An ENN offers the formulation of the predictions of NNs as subjective opinions and learns the function by collecting an amount of evidence that can form the subjective opinions by a deterministic NN from data. However, the ENN is trained as a black box without explicitly considering inherent uncertainty in data with their different root causes, such as vacuity (i.e., uncertainty due to a lack of evidence) or dissonance (i.e., uncertainty due to conflicting evidence). By considering the multidimensional uncertainty, we proposed a novel uncertainty-aware evidential NN called WGAN-ENN (WENN) for solving an out-of-distribution (OOD) detection problem. We took a hybrid approach that combines Wasserstein Generative Adversarial Network (WGAN) with ENNs to jointly train a model with prior knowledge of a certain class, which has high vacuity for OOD samples. Via extensive empirical experiments based on both synthetic and real-world datasets, we demonstrated that the estimation of uncertainty by WENN can significantly help distinguish OOD samples from boundary samples. WENN outperformed in OOD detection when compared with other competitive counterparts.

Disinformation, fake news, and unverified rumors spread quickly in online social networks (OSNs) and manipulate people's opinions and decisions about life events. The solid mathematical solutions of the strategic decisions in OSNs have been provided under game theory models, including multiple roles and features. This work proposes a game-theoretic opinion framework to model subjective opinions and behavioral strategies of attackers, users, and a defender. The attackers use information deception models to disseminate disinformation. We investigate how different game-theoretic opinion models of updating people's subject opinions can influence a way for people to handle disinformation. We compare the opinion dynamics of the five different opinion models (i.e., uncertainty, homophily, assertion, herding, and encounter-based) where an opinion is formulated based on Subjective Logic that offers the capability to deal with uncertain opinions. Via our extensive experiments, we observe that the uncertainty-based opinion model shows the best performance in combating disinformation among all in that uncertainty-based decisions can significantly help users believe true information more than disinformation.

In this work, we proposed a Deep reinforcement learning (DRL)-based NETwork Adaptations for network Resilience algorithm, namely DeepNETAR, which aims to generate robust network topologies against epidemic attacks. In DeepNETAR, a DRL agent aims to generate a robust network topology against epidemic attacks by removing vulnerable edges or adding the least vulnerable edges, given multiple objectives of system security and performance. Most existing network topology adaptation algorithms have used the size of the giant component (SGC) to ensure service availability based on network connectivity. However, in real communication networks, where packets may be dropped either from the presence of inside attackers or congestion on long routes, a larger SGC does not necessarily ensure higher service availability. In addition, for the DRL agent to learn fast and handle multiple, conflicting system objectives, we considered vulnerability-based selection of adaptable edge candidates, fractal-based solution search, and diverse reward functions aiming to achieve multi-objective optimization. Via extensive simulation experiments, we analyzed what DeepNETAR-based schemes using different objectives can achieve those two conflicting system objectives and comparing existing and baseline counterparts.

In this work, we propose a PRivAcy-Diversity-Aware Team Formation framework, namely PRADA-TF, that can be deployed based on the trust relationships between users in online social networks (OSNs). Our proposed PRADA-TF is mainly designed to reflect team members' domain expertise and privacy preserving preferences when a task requires a wide range of diverse domain expertise for its successful completion. The proposed PRADA-TF aims to form a team for maximizing its productivity based on members' characteristics in their diversity, privacy preserving, and information sharing. We leveraged a game theory called Mechanism Design in order for a mechanism designer as a team leader to select team members that can maximize the team's social welfare, which is the sum of all team members' utilities considering team productivity, members' privacy preserving, and potential privacy loss caused by information sharing. To screen a set of candidate teams in the OSN, we built an expert social network based on real coauthorship datasets (i.e., Netscience) with 1,590 scientists. We used the semi-synthetic datasets to construct a trust network based on a belief model called Subjective Logic and identified trustworthy users as candidate team members. Via our extensive simulation experiments, we compared the seven different TF schemes, including our proposed and existing TF algorithms, and analyzed the key factors that can significantly impact the expected and actual social welfare, expected and actual potential privacy loss, and team diversity of a selected team.

Using unreliable information sources generating conflicting evidence may lead to a large uncertainty, which significantly hurts the decision making process. Recently, many approaches have been taken to integrate conflicting data from multiple sources and/or fusing conflicting opinions from different entities. To explicitly deal with uncertainty, a belief model called Subjective Logic (SL), as a variant of Dumpster-Shafer Theory, has been proposed to represent subjective opinions and to merge multiple opinions by offering a rich volume of fusing operators, which have been used to solve many opinion inference problems in trust networks. However, the operators of SL are known to be lack of scalability in inferring unknown opinions from large network data as a result of the sequential procedures of merging multiple opinions. In addition, SL does not consider deriving opinions in the presence of conflicting evidence. In this work, we propose a hybrid inference method that combines SL and Probabilistic Soft Logic (PSL), namely, Collective Subjective Plus, CSL + , which is resistible to highly conflicting evidence or a lack of evidence. PSL can reason a belief in a collective manner to deal with large-scale network data, allowing high scalability based on relationships between opinions. However, PSL does not consider an uncertainty dimension in a subjective opinion. To take benefits from both SL and PSL, we proposed a hybrid approach called CSL + for achieving high scalability and high prediction accuracy for unknown opinions with uncertainty derived from a lack of evidence and/or conflicting evidence. Through the extensive experiments on four semi-synthetic and two real-world datasets, we showed that the CSL + outperforms the state-of-the-art belief model (i.e., SL), probabilistic inference models (i.e., PSL, CSL), and deep learning model (i.e., GCN-VAE-opinion) in terms of prediction accuracy, computational complexity, and real running time.

By leveraging the principle of software polyculture to ensure security in a network, we propose a vulnerability-based software diversity metric to determine how a network topology can be adapted to minimize security vulnerability while maintaining maximum network connectivity. Our proposed metric estimates the software diversity of the node using the vulnerabilities of software packages installed on nearby nodes on attack paths reachable to the node. Our software diversity-based adaptation (SDA) scheme employs the diversity of each node for edge adaptations. These adaptations include the removal of edges that expose high security vulnerability as well as the potential addition of edges between certain nodes with low vulnerabilities associated with them. To validate the proposed SDA scheme, we conduct extensive experiments comparing our approach with counterpart baseline schemes in real networks. Our simulation results demonstrate that SDA outperforms these existing counterparts. We discuss insights into these findings in terms of the effectiveness and efficiency of the proposed SDA scheme under three real network topologies with vastly different network densities.

This work proposes a context-aware trust estimation scheme that can allow roadside units in a vehicular edge network to provide real-time crowdsensing services in a reliable manner by selectively using information from trustworthy sources. Our proposed scheme is novel in that its trust estimation does not require any prior knowledge towards vehicles on roads but quickly obtains an accurate trust value of each vehicle. To that end, we particularly leverage the concept of I-sharing which removes a cold-start problem during the system bootstrapping period. Based on an extensive simulation study, we prove that the proposed scheme outperforms its competitive counterpart and baseline models in terms of trust bias and malicious vehicle detection accuracy.

Thanks to graph neural networks (GNNs), semi-supervised node classification has shown the state-of-the-art performance in graph data. However, GNNs have not considered different types of uncertainties associated with class probabilities to minimize risk of increasing misclassification under uncertainty in real life. In this work, we propose a multi-source uncertainty framework using a GNN that reflects various types of predictive uncertainties in both deep learning and belief/evidence theory domains for node classification predictions. By collecting evidence from the given labels of training nodes, the Graph-based Kernel Dirichlet distribution Estimation (GKDE) method is designed for accurately predicting node-level Dirichlet distributions and detecting out-of-distribution (OOD) nodes. We validated the outperformance of our proposed model compared to the state-of-the-art counterparts in terms of misclassification detection and OOD detection based on six real network datasets. We found that dissonance-based detection yielded the best results on misclassification detection while vacuity-based detection was the best for OOD detection. To clarify the reasons behind the results, we provided the theoretical proof that explains the relationships between different types of uncertainties considered in this work.

Abstract Simple and efficient exploration remains a core challenge in deep reinforcement learning. While many exploration methods can be applied to high-dimensional tasks, these methods manually adjust exploration parameters according to domain knowledge. This paper proposes a novel method that can automatically balance exploration and exploitation, as well as combine on-policy and off-policy update targets through a dynamic weighted way based on value difference. The proposed method does not directly affect the probability of a selected action but utilizes the value difference produced during the learning process to adjust update target for guiding the direction of agent’s learning. We demonstrate the performance of the proposed method on CartPole-v1, MountainCar-v0, and LunarLander-v2 classic control tasks from the OpenAI Gym. Empirical evaluation results show that by integrating on-policy and off-policy update targets dynamically, this method exhibits superior performance and stability than does the exclusive use of the update target.

We propose a lightweight specification-based misbehavior detection management technique to efficiently and effectively detect misbehavior of an IoT device embedded in a medical cyber physical system through automatic model checking and formal verification. We verify our specification-based misbehavior detection technique with a patient-controlled analgesia (PCA) device embedded in a medical health monitoring system. Through extensive ns3 simulation, we verify its superior performance over popular machine learning anomaly detection methods based on support vector machine (SVM) and k-nearest neighbors (KNN) techniques in both effectiveness and efficiency performance metrics.

Moving target defense (MTD) has emerged as a proactive defense mechanism aiming to thwart a potential attacker. The key underlying idea of MTD is to increase uncertainty and confusion for attackers by changing the attack surface (i.e., system or network configurations) that can invalidate the intelligence collected by the attackers and interrupt attack execution; ultimately leading to attack failure. Recently, the significant advance of software-defined networking (SDN) technology has enabled several complex system operations to be highly flexible and robust; particularly in terms of programmability and controllability with the help of SDN controllers. Accordingly, many security operations have utilized this capability to be optimally deployed in a complex network using the SDN functionalities. In this paper, by leveraging the advanced SDN technology, we developed an attack graph-based MTD technique that shuffles a host’s network configurations (e.g., MAC/IP/port addresses) based on its criticality, which is highly exploitable by attackers when the host is on the attack path(s). To this end, we developed a hierarchical attack graph model that provides a network’s vulnerability and network topology, which can be utilized for the MTD shuffling decisions in selecting highly exploitable hosts in a given network, and determining the frequency of shuffling the hosts’ network configurations. The MTD shuffling with a high priority on more exploitable, critical hosts contributes to providing adaptive, proactive, and affordable defense services aiming to minimize attack success probability with minimum MTD cost. We validated the out performance of the proposed MTD in attack success probability and MTD cost via both simulation and real SDN testbed experiments.

Inference of unknown opinions with uncertain, adversarial (e.g., incorrect or conflicting) evidence in large datasets is not a trivial task. Without proper handling, it can easily mislead decision making in data mining tasks. In this work, we propose a highly scalable opinion inference probabilistic model, namely Adversarial Collective Opinion Inference (Adv-COI), which provides a solution to infer unknown opinions with high scalability and robustness under the presence of uncertain, adversarial evidence by enhancing Collective Subjective Logic (CSL) which is developed by combining SL and Probabilistic Soft Logic (PSL). The key idea behind the Adv-COI is to learn a model of robust ways against uncertain, adversarial evidence which is formulated as a min-max problem. We validate the out-performance of the Adv-COI compared to baseline models and its competitive counterparts under possible adversarial attacks on the logic-rule based structured data and white and black box adversarial attacks under both clean and perturbed semi-synthetic and real-world datasets in three real world applications. The results show that the Adv-COI generates the lowest mean absolute error in the expected truth probability while producing the lowest running time among all.

Moving target defense (MTD) is a proactive defense mechanism of changing the attack surface to increase an attacker's confusion and/or uncertainty, which invalidates its intelligence gained through reconnaissance and/or network scanning attacks. In this work, we propose software-defined networking (SDN)-based MTD technique using the shuffling of IP addresses and port numbers aiming to obfuscate both network and transport layers' real identities of the host and the service for defending against the network reconnaissance and scanning attacks. We call our proposed MTD technique Random Host and Service Multiplexing, namely RHSM. RHSM allows each host to use random, multiple virtual IP addresses to be dynamically and periodically shuffled. In addition, it uses short-lived, multiple virtual port numbers for an active service running on the host. Our proposed RHSM is novel in that we employ multiplexing (or de-multiplexing) to dynamically change and remap from all the virtual IPs of the host to the real IP or the virtual ports of the services to the real port, respectively. Via extensive simulation experiments, we prove how effectively and efficiently RHSM outperforms a baseline counterpart (i.e., a static network without RHSM) in terms of the attack success probability and defense cost.

We study how emotions influence norm outcomes in decision-making contexts. Following the literature, we provide baseline Dynamic Bayesian models to capture an agent's two perspectives on a directed norm. Unlike the literature, these models are holistic in that they incorporate not only norm outcomes and emotions but also trust and goals. We obtain data from an empirical study involving game play with respect to the above variables. We provide a step-wise process to discover two new Dynamic Bayesian models based on maximizing log-likelihood scores with respect to the data. We compare the new models with the baseline models to discover new insights into the relevant relationships. Our empirically supported models are thus holistic and characterize how emotions influence norm outcomes better than previous approaches.

As connected and autonomous vehicle technology evolves, the design of in-vehicle network architecture, which connects multiple electronic control units (ECUs) and internal sensors and supports connectivity to the outside of the vehicle, has increased significantly to meet security needs. However, the heterogeneous structure of the in-vehicle network and lack of security consideration has introduced a lack of scalability and security concerns. In this work, we propose a shuffling-based moving target defense (MTD) technique aiming to disturb network reconnaissance attacks and deployed it in the proposed software-defined networking (SDN)-based in-vehicle network architecture. To validate the proposed MTD, we compare the service availability of our proposed MTD and non-MTD counterpart in the presence of the reconnaissance-based false message injection attacks.

In cyberspace, evolutionary strategies are commonly used by both attackers and defenders. For example, an attacker's strategy often changes over the course of time, as new vulnerabilities are discovered and/or mitigated. Similarly, a defender's strategy changes over time. These changes may or may not be in direct response to a change in the opponent's strategy. In any case, it is important to have a set of quantitative metrics to characterize and understand the effectiveness of attackers' and defenders' evolutionary strategies, which reflect their cyber agility. Despite its clear importance, few systematic metrics have been developed to quantify the cyber agility of attackers and defenders. In this paper, we propose the first metric framework for measuring cyber agility in terms of the effectiveness of the dynamic evolution of cyber attacks and defenses. The proposed framework is generic and applicable to transform any relevant, quantitative, and/or conventional static security metrics (e.g., false positives and false negatives) into dynamic metrics to capture dynamics of system behaviors. In order to validate the usefulness of the proposed framework, we conduct case studies on measuring the evolution of cyber attacks and defenses using two real-world datasets. We discuss the limitations of the current work and identify future research directions.

In the network science domain, a larger size of the giant component (i.e., the largest cluster of nodes) represents higher network resilience in terms of maximizing network availability in the presence of attacks. However, this does not necessarily represent how well the network provides promised services under attacks and/or failures. We aim to improve network resilience by introducing network adaptability (i.e., reconfiguration of a network topology), in addition to fault-tolerance. We develop a suite of strategies adopting processes from percolation theory, describing the process to percolate into a medium, for a tactical, mission-oriented network. This network is service-oriented, characterized by a number of task teams where each resource-restricted node aims to maximize resource utilization while completing multiple tasks without failure. We investigate how node failures can trigger overloads, leading to cascading failures. We consider various attack behaviors (infectious, non-infectious, random, or targeted) and analyze their effects. Through extensive simulations, we show the outperformance of the proposed adaptation strategy compared with the performance of the existing counterparts in terms of the size of the giant component, the utilization of resources, the number of alive task teams (or mission success ratio), and the adaptation cost for a large-scale, mission-oriented network under attack.

Subjective Logic (SL) is one of well-known belief models that can explicitly deal with uncertain opinions and infer unknown opinions based on a rich set of operators of fusing multiple opinions. Due to high simplicity and applicability, SL has been substantially applied in a variety of decision making in the area of cybersecurity, opinion models, trust models, and/or social network analysis. However, SL and its variants have exposed limitations in predicting uncertain opinions in real-world dynamic network data mainly in three-fold: (1) a lack of scalability to deal with a large-scale network; (2) limited capability to handle heterogeneous topological and temporal dependencies among node-level opinions; and (3) a high sensitivity with conflicting evidence that may generate counterintuitive opinions derived from the evidence. In this work, we proposed a novel deep learning (DL)-based dynamic opinion inference model while node-level opinions are still formalized based on SL meaning that an opinion has a dimension of uncertainty in addition to belief and disbelief in a binomial opinion (i.e., agree or disagree). The proposed DL-based dynamic opinion inference model overcomes the above three limitations by considering the following: (1) state-of-the-art DL techniques, such as the Graph Convolutional Network (GCN) and the Gated Recurrent Units (GRU), for modeling the topological and temporal heterogeneous dependency information of a given dynamic network; (2) modeling conflicting opinions based on robust statistics; and (3) a highly scalable inference algorithm to predict dynamic, uncertain opinions in a linear computation time. We validated the outperformance of our proposed DL-based algorithm (i.e., GCN-GRU-opinion model) via extensive comparative performance analysis based on a real-world dataset.

In LinkedIn, skill endorsements are used as a key indicator to assess a user's skill level so that a recruiter can find right candidates. However, the simple counting of skill endorsements has shown serious limitations to correctly assess a user's skill level particularly when false endorsements prevail for self-promotion or collusive promotion. To address this issue, we propose a framework using regression analysis to evaluate assessment methods under varying the degree of false endorsements. We define two types of assessment scores: authority score (AS) and credit score (CS). The AS is calculated based on an endorser's expertise level in a particular skill set while the CS is estimated based on the endorser's trustworthiness. Based on these two types of skill assessments, we define a hybrid method that calculates AS based on CS, which is called credit-based AS, or namely CAS. We conduct performance analysis by comparing the AS and CAS with the endorsement count (EC) as a baseline model, which is currently used in LinkedIn. Through the extensive regression analysis, under varying the degree of false endorsements from 10% to 25%, we observe that the CAS outperforms AS and EC with less than 10% of false endorsements. On the other hand, the AS performs the best among three methods with 10%-25% of false endorsements. Based on the regression analysis of the skillset assessment framework, the real dataset collected for this study is likely to have 4%-6% of false endorsements. This finding implies that an endorser's expertise appears more critical than the endorser's trustworthiness under more presence of false endorsements because the unqualified endorser's endorsement may not be necessarily true.

In this work, we study different types of uncertainty in subjective opinions based on the internal belief mass distribution and the base rate distribution. Subjective opinions which are used as arguments in subjective logic (SL) expand the traditional belief functions by including base rate distributions. Fundamental uncertainty characteristics of a given opinion depend on its `singularity', `vagueness', `vacuity', `dissonance', `consonance' and `monosonance'. We define those concepts in the formalism of SL and show how these characteristics can be manifested in the three different opinion classes which are binomial, multinomial, and hyper-opinions. We clarify the relationships between the uncertainty characteristics and discuss how they influence decision making in SL.

The proliferation of information disseminated by public/social media has made decision-making highly challenging due to the wide availability of noisy, uncertain, or unverified information. Although the issue of uncertainty in information has been studied for several decades, little work has investigated how noisy (or uncertain) or valuable (or credible) information can be formulated into people's opinions, modeling uncertainty both in the quantity and quality of evidence leading to a specific opinion. In this work, we model and analyze an opinion and information model by using Subjective Logic where the initial set of evidence is mixed with different types of evidence (i.e., pro vs. con or noisy vs. valuable) which is incorporated into the opinions of original propagators, who propagate information over a network. With the help of an extensive simulation study, we examine how the different ratios of information types or agents' prior belief or topic competence affect the overall information diffusion. Based on our findings, agents' high uncertainty is not necessarily always bad in making a right decision as long as they are competent enough not to be at least biased towards false information (e.g., neutral between two extremes).

Subjective Logic (SL) is one of well-known belief models that can explicitly deal with uncertain opinions and infer unknown opinions based on a rich set of operators of fusing multiple opinions. Due to high simplicity and applicability, SL has been popularly applied in a variety of decision making in the area of cybersecurity, opinion models, and/or trust / social network analysis. However, SL has been facing an issue of scalability to deal with a large-scale network data. In addition, SL has shown a bounded prediction accuracy due to its inherent parametric nature by treating heterogeneous data and network structure homogeneously based on the assumption of a Bayesian network. In this work, we take one step further to deal with uncertain opinions for unknown opinion inference. We propose a deep learning (DL)-based opinion inference model while node-level opinions are still formalized based on SL. The proposed DL-based opinion inference model handles node-level opinions explicitly in a large-scale network using graph convoluational network (GCN) and variational autoencoder (VAE) techniques. We adopted the GCN and VAE due to their powerful learning capabilities in dealing with a large-scale network data without parametric fusion operators and/or Bayesian network assumption. This work is the first that leverages the merits of both DL (i.e., GCN and VAE) and a belief model (i.e., SL) where each node level opinion is modeled by the formalism of SL while GCN and VAE are used to achieve non-parametric learning with low complexity. By mapping the node-level opinions modeled by the GCN to their equivalent Beta PDFs (probability density functions), we develop a network-driven VAE to maximize prediction accuracy of unknown opinions while significantly reducing algorithmic complexity. We validate our proposed DL-based algorithm using real-world datasets via extensive simulation experiments for comparative performance analysis.

Percolation theory has been studied to investigate network resilience by identifying a critical value of node occupation probability where a giant component (i.e., a largest component in a network) represents high network resilience. However, the concept of network resilience studied in percolation theory has been limited to measuring network fault-tolerance in the presence of failures/attacks. In this work, we take a step to extend the concept of network resilience beyond fault-tolerance by introducing network adaptability. We consider a tactical network where a node is executing multiple tasks by belonging to multiple task groups. In this type of tactical networks, a node is limited with its resource while aiming to maximize its resource utilization and task execution without being overloaded. We investigate network resilience and resource utilization of the tactical network where the network is attacked by either infectious or non-infectious attacks. To mitigate the impact of the failed/attacked nodes in the network, we propose a suite of the proposed adaptation strategies to deal with correlated, cascading failures caused by overloaded nodes due to increased workload introduced by other failed nodes. We conduct a comparative performance analysis of the set of proposed adaptation strategies and a baseline scheme with no adaptation based on metrics including a size of a giant component, node resource utilization, a number of active tasks in execution, and adaptation cost. Our simulation results show that a large size of a giant component does not necessarily ensure high resource utilization of nodes and task performance in the given tactical network.

Unmanned Aerial Vehicles (UAV)-based civilian or military applications become more critical to serving civilian and/or military missions. The significantly increased attention on UAV applications also has led to security concerns particularly in the context of networked UAVs. Networked UAVs are vulnerable to malicious attacks over open-air radio space and accordingly intrusion detection systems (IDSs) have been naturally derived to deal with the vulnerabilities and/or attacks. In this paper, we briefly survey the state-of-the-art IDS mechanisms that deal with vulnerabilities and attacks under networked UAV environments. In particular, we classify the existing IDS mechanisms according to information gathering sources, deployment strategies, detection methods, detection states, IDS acknowledgment, and intrusion types. We conclude this paper with research challenges, insights, and future research directions to propose a networked UAVIDS system which meets required standards of effectiveness and efficiency in terms of the goals of both security and performance.

This article discusses a lightweight behavior rule specification-based monitoring solution for identifying misbehavior of an embedded IoT device. These unusual activities are exhibited because of attacks exploiting the vulnerability exposed through automatic model checking and formal verification. It is conclusive in the presented research that rule specification-based misbehavior detection technique outperforms contemporary anomaly-based misbehavior detection techniques for an unmanned aerial vehicle (UAV) cyber-physical system.

Smart homes (SHs) aim at forming an energy optimized environment that can efficiently regulate the use of various Internet of Things (IoT) devices in its network. Real-time electricity pricing models along with SHs provide users an opportunity to reduce their electricity expenditure by responding to the pricing that varies with different times of the day, resulting in reducing the expenditure at both customers’ and utility provider’s end. However, responding to such prices and effectively scheduling the appliances under such complex dynamics is a challenging optimization problem to be solved by the provider or by third party services. As communication in SH-IoT environment is extremely sensitive and private, reporting of such usage information to the provider to solve the optimization has a potential risk that the provider or third party services may track users’ energy consumption profile which compromises users’ privacy. To address these issues, we developed a homomorphic encryption-based alternating direction method of multipliers approach to solve the cost-aware appliance scheduling optimization in a distributed manner and schedule home appliances without leaking users’ privacy. Through extensive simulation study considering real-world datasets, we show that the proposed secure appliance scheduling for flexible and efficient energy consumption scheme, namely SAFE, effectively lowers electricity cost while preserving users’ privacy.

In this paper, we study the evolution of opinions where people are not sure of their own opinions and/or their opinions may be conflicting to others' in social networks. We model two types of agents so-called informed agents (IAs) and uninformed agents (UIAs). The IAs have a strong opinion agreeing or disagreeing toward a proposition without being influenced by other agents' opinions and have a high confidence (low uncertainty) toward its own opinion. The UIAs have a weak opinion without either agreeing or disagreeing toward a proposition and lack confidence with a high uncertainty. Based on subjective logic, we consider a binomial opinion to deal with an opinion with a degree of uncertainty. We develop two types of trust attitudes for agents to update their opinions upon their interactions with other agents: uncertainty-based trust (UT) and similarity-based trust (ST). In the UT, a UIA updates its opinion based on an interacting agent's uncertainty toward a proposition. In the ST, a UIA updates its opinion based on the degree of similarity between its own opinion and the interacting agent's opinion toward the proposition. Our results show that more IAs slow down the convergence of the opinions under the UT while they can quickly lead to opinion convergence under the ST. In addition, the ST leads uncertain opinions to two extremes, either 0 or 1, if consensus exists. On the other hand, the UT can make opinions converge to a certain point between two extreme opinions although the converged point is significantly affected by the dominant agents' opinions. Furthermore, we observe that under the UT, more IAs with a high centrality increase dissonance of opinions, while more IAs with a low centrality offer better chances for opinion consensus in both the UT and the ST.

Delay tolerant networks (DTNs) are often encountered in military network environments where end-to-end connectivity is not guaranteed due to frequent disconnection or delay. This work proposes a provenance-based trust framework, namely PROVEST (PROVEnance-baSed Trust model) that aims to achieve accurate peer-to-peer trust assessment and maximize the delivery of correct messages received by destination nodes while minimizing message delay and communication cost under resource-constrained network environments. Provenance refers to the history of ownership of a valued object or information. We leverage the interdependency between trustworthiness of information source and information itself in PROVEST. PROVEST takes a data-driven approach to reduce resource consumption in the presence of selfish or malicious nodes while estimating a node's trust dynamically in response to changes in the environmental and node conditions. This work adopts a model-based method to evaluate the performance of PROVEST (i.e., trust accuracy and routing performance) using Stochastic Petri Nets. We conduct a comparative performance analysis of PROVEST against existing trust-based and non-trust-based DTN routing protocols to analyze the benefits of PROVEST. We validate PROVEST using a real dataset of DTN mobility traces.

The accurate measurement of security metrics is a critical research problem, because an improper or inaccurate measurement process can ruin the usefulness of the metrics. This is a highly challenging problem, particularly when the ground truth is unknown or noisy. In this paper, we measure five malware detection metrics in the absence of ground truth, which is a realistic setting that imposes many technical challenges. The ultimate goal is to develop principled, automated methods for measuring these metrics at the maximum accuracy possible. The problem naturally calls for investigations into statistical estimators by casting the measurement problem as a statistical estimation problem. We propose statistical estimators for these five malware detection metrics. By investigating the statistical properties of these estimators, we characterize when the estimators are accurate, and what adjustments can be made to improve them under what circumstances. We use synthetic data with known ground truth to validate these statistical estimators. Then, we employ these estimators to measure five metrics with respect to a large data set collected from VirusTotal.

We propose a context-aware trust management model called CATrust for service-oriented ad hoc networks such as peer-to-peer and Internet of Things networks wherein a node can be a service requester or a service provider. The novelty of our design lies in the use of logistic regression to dynamically estimate trustworthiness of a service provider based on its service behavior patterns in response to context environment changes. We develop a recommendation filtering mechanism to effectively screen out dishonest recommendations even in extremely hostile environments in which the majority recommenders are dishonest. We demonstrate desirable convergence, accuracy, and resiliency properties of CATrust. We also demonstrate that CATrust outperforms contemporary peer-to-peer and Internet of Things trust models in terms of service trust prediction accuracy against collusion recommendation attacks.

Subjective Logic (SL), as one of the state-of-the-art belief models, has been proposed to model an opinion that explicitly deals with its uncertainty. SL offers a variety of operators to update opinions consisting of belief, disbelief, and uncertainty. However, SL operators lack scalability to derive opinions from a large-scale network data due to the sequential procedures of combining two opinions, instead of collective procedures dealing with multiple opinions concurrently. In addition, SL's performance in predicting unknown opinions has been validated only when the uncertainty mass is sufficiently low. To enhance scalability and prediction accuracy of unknown opinions in SL, we take a hybrid approach by combining SL with Probabilistic Soft Logic (PSL). PSL provides collective reasoning with high scalability based on relationships between opinions but does not deal with uncertainty. By taking the merits of both SL and PSL, we propose a probabilistic logic algorithm, called Collective Subjective Logic (CSL) that provides high scalability and high prediction accuracy while dealing with uncertain opinions. Our proposed CSL is generic to deal with uncertain opinions with both high scalability and high prediction accuracy of unknown opinions over a large-scale network dataset. Through the extensive simulation experiments, we validated the outperformance of CSL compared against SL and PSL in terms of prediction accuracy of unknown opinions and algorithmic complexity using Epinions and two road traffic datasets.

To stop or mitigate the dissemination of false information in social networks, many studies have investigated the minimum number of seeding nodes required to significantly reduce the impact of false information. Although a person's confidence level, such as perceived certainty, and/or prior belief towards a given proposition can significantly affect their decision of whether to believe in true or false information, these topics have not been studied to date. In this work, we propose an opinion model based on Subjective Logic (SL), defining an opinion in belief, disbelief, and uncertainty, to study how to eradicate or mitigate the impact of false information by propagating true information to counter it. In the current form of an opinion in SL, when two agents interact with each other and update their opinions based on a consensus operator offered by SL, uncertainty continuously reduces whenever any new information, even conflicting evidence, is received. However, in reality, if a body of evidence is conflicting, with equal amounts supporting opposite positions, people often tend to be confused, leading to higher level of uncertainty. We enhance SL to deal with conflicting information associated with uncertainty. We map agents' opinion composition into each state in the SIR (Susceptible-Infected- Recovered) model to estimate the proportion of recovered agents who believe in true information. Our results show that agents' prior belief unfavoring false information can help guide their decisions towards a belief in true information even under high uncertainty. Further, having more true informers in a network can significantly increase the number of agents who believe in true information and the effect is more pronounced than having more frequent propagation of true information by fewer true informers.

Linking multiple accounts owned by the same user across different online social networks (OSNs) is an important issue in social networks, known as identity reconciliation. Graph matching is one of popular techniques to solve this problem by identifying a map that matches a set of vertices across different OSNs. Among them, percolation-based graph matching (PGM) has been explored to identify entities belonging to a same user across two different networks based on a set of initial pre-matched seed nodes and graph structural information. However, existing PGM algorithms have been applied in only undirected networks while many OSNs are represented by directional relationships (e.g., followers or followees in Twitter or Facebook). For PGM to be applicable in real world OSNs represented by directed networks with a small set of overlapping vertices, we propose a percolation-based directed graph matching algorithm, namely PDGM, by considering the following two key features: (1) similarity of two nodes based on directional relationships (i.e., outgoing edges vs. incoming edges); and (2) celebrity penalty such as penalty given for nodes with a high in-degree. Through the extensive simulation experiments, our results show that the proposed PDGM outperforms the baseline PGM counterpart that does not consider either directional relationships or celebrity penalty.

We propose and analyze a trust management protocol in service-oriented mobile ad hoc networks (MANETs) populated with service providers and service requesters, and demonstrate the resiliency and convergence properties against bad-mouthing, ballot-stuffing, opportunistic service, and self-promotion attacks. To demonstrate the applicability, we consider a mission-driven service-oriented MANET that must handle dynamically arriving tasks to achieve multiple conflicting objectives. We devise a trust-based heuristic algorithm based on auctioning with local knowledge of node status to solve this node-to-task assignment problem with multiobjective optimization (MOO) requirements. Our trust-based heuristic algorithm has a polynomial runtime complexity, rather than an exponential runtime complexity as in existing work, thus allowing dynamic node-to-task assignment to be performed at runtime. It outperforms a nontrust-based counterpart using blacklisting techniques while performing close to the ideal solution quality with perfect knowledge of node status over a wide range of environmental conditions. We conduct extensive sensitivity analysis of the results with respect to key design parameters and alternative trust protocol designs. We also develop a table-lookup method to apply the best trust protocol parameter settings upon detection of dynamically changing environmental conditions to maximize MOO performance.

With the proliferation of fairly powerful mobile devices and ubiquitous wireless technology, we see a transformation from traditional mobile ad hoc networks (MANETs) into a new era of service-oriented MANETs wherein a node can provide and receive services. Requested services must be decomposed into more abstract services and then bound; we formulate this as a multi-objective optimization (MOO) problem to minimize the service cost, while maximizing the quality of service and quality of information in the service a user receives. The MOO problem is an SP-to-service assignment problem. We propose a multidimensional trust based algorithm to solve the problem. We carry out an extensive suite of simulations to test the relative performance of the proposed trust-based algorithm against a non-trust-based counterpart and an existing single-trust-based beta reputation scheme. Our proposed algorithm effectively filters out malicious nodes exhibiting various attack behaviors by penalizing them with loss of reputation, which ultimately leads to high user satisfaction. Further, our proposed algorithm is efficient with linear runtime complexity while achieving a close-to-optimal solution.

In online social networks (OSNs), individual users have a strong desire to expand their social networks through OSN activities and try to maximize the benefits from the social relationships, called social capital. However, with a large-scale social network, their privacy rights have been significantly intruded by adversarial users that perform social attacks including false / illegal private information dissemination or the use of fake identities. In this work, we study how individual users can expand their social networks by making trustworthy friends while not leaking their private information out to unauthorized parties or social attackers. We adopt the concepts of trust and reputation in order to preserve users' privacy while enhancing their social capital in OSNs. Given a social network topology from the Facebook, we model an individual user's interactions with other users based on feeding (e.g., posting information) and feedback behaviors (e.g., providing likes or comments). Our results show that there exists a tradeoff between social capital and privacy preservation. In addition, we show there exists a balance point of social capital and privacy thresholds that maximizes correct information diffusion while minimizing illegal private information leakout, given users' risk appetite for preserving privacy.

Due to the proliferation of using various online social media, individual users put their privacy at risk by posting and exchanging enormous amounts of messages or activities with other users. This causes a serious concern about leaking private information out to malevolent entities without users’ consent. This work proposes a reputation model in order to achieve efficient and effective privacy preservation in which a user’s reputation score can be used to set the level of privacy and accordingly to determine the level of visibility for all messages or activities posted by the users. We derive a user’s reputation based on both individual and relational characteristics in online social network environments. We demonstrate how the proposed reputation model can be used for automatic privacy assessment and accordingly visibility setting for messages / activities created by a user.

We propose and analyze a trust management protocol for autonomous service-oriented mobile ad hoc networks (MANETs) populated with service providers (SPs) and service requesters (SRs). We demonstrate the resiliency and convergence properties of our trust protocol design for service-oriented MANETs in the presence of malicious nodes performing opportunistic service attacks and slandering attacks. Further, we consider a situation in which a mission comprising dynamically arriving tasks must achieve multiple conflicting objectives, including maximizing the mission reliability, minimizing the utilization variance, and minimizing the delay to task completion. We devise a trust-based heuristic algorithm to solve this multi-objective optimization problem with a linear runtime complexity, thus allowing dynamic node-to-task assignment to be performed at runtime. Through extensive simulation, we demonstrate that our trust-based node-to-task assignment algorithm outperforms a non-trust-based counterpart using blacklisting techniques while performing close to the ideal solution quality with perfect knowledge of node reliability over a wide range of environmental conditions.

In a military tactical network, maintaining trust among members in a mission group is critical to successful mission completion. However, maintaining high trust among group members in a resource-restricted tactical environment detrimentally reduces system lifetime, which may lead to mission failure or low mission effectiveness. In this paper, we aim to investigate the relationships between group trust and system lifetime [i.e., survivability measuring mean time to mission failure (MTTMF)] and to capture mission effectiveness achieved by the mission group based on the tradeoff between these two goals. We employ a composite trust capturing various angles of trust concept derived from communication, information, and social networks. We take a game theoretic approach using the so called Aoyagi's game theory, enforcing nodes to exhibit desirable behavior based on reward or penalty given by the system. In designing reward/penalty mechanisms, we adopt the concept of aspiration level, defining success or failure based on a goal set by the system, and prove there exists an optimal trust threshold maximizing both MTTMF (i.e., system lifetime/survivability) and group trust. We devised a mission effectiveness metric based on both the metrics having conflicting goals. We developed an analytical model using Stochastic Petri Nets, and validated the analytical results with simulation results. We conducted comparative performance analyzes of the variations of the proposed scheme with respect to a node's decision nature (i.e., rational versus altruistic) and trust threshold policy (static versus dynamic) in resource-constrained tactical environments.

In a military tactical network where a trust authority (e.g., a commander) makes a decision during a mission, assessing the trustworthiness of participating entities accurately is critical to mission success. In this work, we propose a trust-based reputation management scheme, called GlobalTrust, for minimizing false decisions on the reputation of nodes in the network. In the proposed scheme, nodes may be compromised and provide incorrect opinions to the trust authority, who conducts reputation evaluation towards all nodes based on the provided opinions. GlobalTrust achieves three goals: (1) maintaining a consistent global view towards each node; (2) obtaining high resiliency against various attack patterns; and (3) attaining highly accurate reputation values of nodes. Through extensive simulations comparing GlobalTrust with other existing schemes, we show that GlobalTrust minimizes false decisions while maintaining high resilience against various attack behaviors. Specifically, under various attacks, GlobalTrust can achieve a highly accurate consistent view on nodes' reputations even when the number of malicious nodes is up to 40% of all participating nodes.

Delay tolerant networks (DTNs) are characterized by high end-to-end latency, frequent disconnection, and opportunistic communication over unreliable wireless links. In this paper, we design and validate a dynamic trust management protocol for secure routing optimization in DTN environments in the presence of well-behaved, selfish and malicious nodes. We develop a novel model-based methodology for the analysis of our trust protocol and validate it via extensive simulation. Moreover, we address dynamic trust management, i.e., determining and applying the best operational settings at runtime in response to dynamically changing network conditions to minimize trust bias and to maximize the routing application performance. We perform a comparative analysis of our proposed routing protocol against Bayesian trust-based and non-trust based (PROPHET and epidemic) routing protocols. The results demonstrate that our protocol is able to deal with selfish behaviors and is resilient against trust-related attacks. Furthermore, our trust-based routing protocol can effectively trade off message overhead and message delay for a significant gain in delivery ratio. Our trust-based routing protocol operating under identified best settings outperforms Bayesian trust-based routing and PROPHET, and approaches the ideal performance of epidemic routing in delivery ratio and message delay without incurring high message or protocol maintenance overhead.

A temporary coalition is often formed to pursue a common goal based on the collaboration of multiple partners who may have their own objectives. The coalition network must attain multiple objectives, under resource constraints and time deadlines. We propose a task assignment algorithm for a scenario where tasks are dynamic, with different arrival times and deadlines. We propose a heuristic coalition formation technique that uses multiple dimensions of trust (i.e., integrity, competence, social connectedness, and reciprocity) to assess trust of each entity. The proposed scheme enables task leaders to make critical assignment decisions based on assessed trustworthiness of entities. We consider three different objectives, namely, maximizing resilience and resource utilization while minimizing delay to task completion. We devise a ranking-based heuristic with linear runtime complexity to select members based on risk derived from trust assessment of nodes. We validate the performance of our proposed scheme by comparing our scheme with a non-trust-based baseline scheme as well as a global optimal solution implemented with the Integer Linear Programming technique.

We propose a highly scalable cluster-based hierarchical trust management protocol for wireless sensor networks (WSNs) to effectively deal with selfish or malicious nodes. Unlike prior work, we consider multidimensional trust attributes derived from communication and social networks to evaluate the overall trust of a sensor node. By means of a novel probability model, we describe a heterogeneous WSN comprising a large number of sensor nodes with vastly different social and quality of service (QoS) behaviors with the objective to yield "ground truth" node status. This serves as a basis for validating our protocol design by comparing subjective trust generated as a result of protocol execution at runtime against objective trust obtained from actual node status. To demonstrate the utility of our hierarchical trust management protocol, we apply it to trust-based geographic routing and trust-based intrusion detection. For each application, we identify the best trust composition and formation to maximize application performance. Our results indicate that trust-based geographic routing approaches the ideal performance level achievable by flooding-based routing in message delivery ratio and message delay without incurring substantial message overhead. For trust-based intrusion detection, we discover that there exists an optimal trust threshold for minimizing false positives and false negatives. Furthermore, trust-based intrusion detection outperforms traditional anomaly-based intrusion detection approaches in both the detection probability and the false positive probability.

We propose a trust-based intrusion detection scheme utilizing a highly scalable hierarchical trust management protocol for clustered wireless sensor networks. Unlike existing work, we consider a trust metric considering both quality of service (QoS) trust and social trust for detecting malicious nodes. By statistically analyzing peer-to-peer trust evaluation results collected from sensor nodes, each cluster head applies trust-based intrusion detection to assess the trustworthiness and maliciousness of sensor nodes in its cluster. Cluster heads themselves are evaluated by the base station. We develop an analytical model based on stochastic Petri nets for performance evaluation of the proposed trust-based intrusion detection scheme, as well as a statistical method for calculating the false alarm probability. We analyze the sensitivity of false alarms with respect to the minimum trust threshold below which a node is considered malicious. Our results show that there exists an optimal trust threshold for minimizing false positives and false negatives. Further, the optimal trust threshold differs depending on the anticipated wireless sensor network lifetime.

We propose and analyze a class of trust management protocols for encounter-based routing in delay tolerant networks (DTNs). The underlying idea is to incorporate trust evaluation in the routing protocol, considering not only quality-of-service (QoS) trust properties (connectivity) but also social trust properties (honesty and unselfishness) to evaluate other nodes encountered. Two versions of trust management protocols are considered: an equal-weight QoS and social trust management protocol (called trust-based routing) and a QoS only trust management protocol (called connectivity-based routing). By utilizing a stochastic Petri net model describing a DTN behavior, we analyze the performance characteristics of these two routing protocols in terms of message delivery ratio, latency, and message overhead. We also perform a comparative performance analysis with epidemic routing for a DTN consisting of heterogeneous mobile nodes with vastly different social and networking behaviors. The results indicate that trust-based routing approaches the ideal performance of epidemic routing in delivery ratio, while connectivity-based routing approaches the ideal performance in message delay of epidemic routing, especially as the percentage of selfish and malicious nodes present in the DTN system increases. By properly selecting weights associated with QoS and social trust metrics for trust evaluation, our trust management protocols can approximate the ideal performance obtainable by epidemic routing in delivery ratio and message delay without incurring high message overhead.

For mission-oriented mobile group systems designed to continue mission execution in hostile environments in the presence of security attacks, it is critical to properly deploy intrusion detection techniques to cope with insider attacks to enhance the system reliability. In this paper, we analyze the effect of intrusion detection system (IDS) techniques on the reliability of a mission-oriented group communication system consisting of mobile groups set out for mission execution in mobile ad hoc networks. Unlike the common belief that IDS should be executed as often as possible to cope with insider attacks to prolong the system lifetime, we discover that IDS should be executed at an optimal rate to maximize the mean time to failure of the system. Further, the optimal rate at which IDS is executed depends on the operational conditions, system failure definitions, attacker behaviors, and IDS techniques used. We develop mathematical models based on Stochastic Petri nets to identify the optimal rate for IDS execution to maximize the mean time to failure of the system, when given a set of parameter values characterizing the operational conditions, and attacker behaviors.

Under highly security vulnerable, resource restricted, and dynamically changing mobile ad hoc environments, it is critical to be able to maximize the system lifetime while bounding the communication response time for mission-oriented mobile groups. In this paper, we analyze the tradeoff of security versus performance for distributed intrusion detection protocols employed in mobile group communication systems (GCSs). We investigate a distributed voting-based intrusion detection protocol for GCSs in multi-hop mobile ad hoc networks and examine the effect of intrusion detection on system survivability measured by the mean time to security failure (MTTSF) metric and efficiency measured by the communication cost metric. We identify optimal design settings under which the MTTSF metric can be best traded off for the communication cost metric or vice versa.

We develop a class of adaptive security protocols with designs to allow group communication systems (GCSs) in mobile ad hoc networks (MANETs) to dynamically adjust operational settings to best satisfy application-imposed performance and security requirements, leveraging the inherent tradeoff between security and performance properties of the system. These adaptive security protocols include an intrusion detection protocol for dealing with insider attacks and a scalable region-based hierarchical group key management protocol for dealing with outsider attacks. Our design settings include the time interval over which intrusion detection should be performed, and the regional area size for the region-based hierarchical group key management protocol for group key management. When given a set of parameter values characterizing operational and environmental conditions of a GCS, we identify optimal design settings to be used by the system dynamically to maximize the mean time to security failure of the system while minimizing the total group communication cost incurred for GCSs in MANET environments.

In this paper, we analyze the effect of intrusion detection system (IDS) techniques on failure time of a mission-oriented group communication system consisting of mobile groups set out for mission execution in mobile ad hoc networks. Unlike the common belief that IDS should be executed as often as possible to cope with insider attacks to prolong the system lifetime, we discover that IDS should be executed at an optimal rate in order to maximize the system lifetime. Further, the optimal rate at which IDS is executed depends on the operational conditions, system failure definitions, attacker behaviors, and IDS techniques used. We develop mathematical models based on stochastic petri nets to identify the optimal rate for IDS execution to maximize the mean time to security failure of the system, when given a set of parameter values characterizing the operational conditions and attacker behaviors.

To meet the development condition of inside of the country, domestic development methodologies are made with the abbreviation of MaRMI (Magic and Robust Methodology Integrated) in a series of methodologies in South Korea. The MaRMI have the four different methodologies for developing information, object-oriented, component-based, embedded systems. In this paper, the authors describe the feature and structure of each methodology and show our movement for the methodology transfer.