Fake news detection has been a long-standing research focus in social networks. Recent studies suggest that incorporating sentiment information from both news content and user comments can enhance detection performance. However, existing approaches typically treat sentiment features as auxiliary signals, overlooking role differentiation, that is, the same sentiment polarity may originate from users with distinct roles, thereby limiting their ability to capture nuanced patterns for effective detection. To address this issue, we propose SARC, a Sentiment-Augmented Role Clustering framework which utilizes sentiment-enhanced deep clustering to identify user roles for improved fake news detection. The framework first generates user features through joint comment text representation (with BiGRU and Attention mechanism) and sentiment encoding. It then constructs a differentiable deep clustering module to automatically categorize user roles. Finally, unlike existing approaches which take fake news label as the unique supervision signal, we propose a joint optimization objective integrating role clustering and fake news detection to further improve the model performance. Experimental results on two benchmark datasets, RumourEval-19 and Weibo-comp, demonstrate that SARC achieves superior performance across all metrics compared to baseline models. The code is available at: https://github.com/jxshang/SARC.

Signed Graph Neural Networks (SGNNs) have achieved outstanding performance in Link Sign Prediction (LSP), which involves predicting the existence and polarity of edges, by effectively modeling positive and negative interactions in signed graphs. However, their black-box nature raises transparency concerns, necessitating faithful explanations of model behavior to ensure trustworthiness and accountability. Existing eXplainable GNN (XGNN) methods, primarily designed for unsigned graphs, struggle to provide meaningful and human-understandable explanations for SGNN-based LSP, often generating disconnected subgraph explanations or neglecting the unique sign interactions. To address the gap, we propose SGExplainer, a novel method that leverages balanced paths, a concept rooted in signed graph theory, to provide clear and faithful explanations for LSP. SGExplainer employs a path-enforcing mask learning framework that ensures interpretable balanced path generation while maintaining explanation fidelity. Extensive experiments on real-world signed graphs demonstrate that SGExplainer consistently provides faithful and intuitive explanations for various SGNNs, outperforming state-of-the-art baselines in explanation quality, interpretability, and efficiency.

Graph-structured data effectively captures complex relationships in diverse domains such as social networks, financial transactions, citation networks, and recommendation systems. Graph Neural Networks (GNNs) excel in learning intricate topological patterns, yielding strong performance on tasks like node classification and link prediction. However, real-world graphs often scale to millions of nodes and billions of directed edges, posing significant computational and storage challenges for GNN training that frequently exceed available hardware limits. Although graph sampling and distillation techniques alleviate these issues by subsampling or creating surrogate graphs, they primarily handle undirected graphs, neglecting directional semantics that are crucial for applications like fraud detection and causal analysis. To address these limitations, we introduce the Directed Graph Distillation (Dir-GD) framework, which combines distributed learning with community detection to divide large directed graphs into independent subgraphs for distributed directed GNN training. This process culminates in parameter aggregation to produce a compact global synthetic graph that preserves essential topology and directionality. Extensive experiments on large-scale datasets, such as the million-node soc-pokec-relationships, demonstrate over 91% accuracy at 0.001 distillation ratios, accompanied by substantial memory and runtime savings. This work pioneers directed graph distillation as a key paradigm for analyzing ultra-large directed graphs, offering a scalable solution that maintains high fidelity in compressed representations.

Quasi-clique is one of the most fundamental models for characterizing cohesive subgraphs in network analysis. However, existing quasi-clique definitions and identification algorithms are designed for unsigned graphs, while many real-world networks are modeled as signed graphs with positive and negative edges representing cooperative and adversarial interactions between entities. Therefore, it remains an open problem to define a quasi-clique model tailored for signed graphs. Motivated by this, we propose the maximal balanced \( (\gamma_{1},\gamma_{2}) \) -quasi-clique (MBQC) model, which not only preserves the essence of quasi-completeness but also aligns with the foremost structural balance theory for signed graphs. Specifically, we formulate the problem of MBQCs enumeration in a given signed graph and prove its NP-hardness. To address this problem, we devise a novel branch-and-bound algorithm to efficiently enumerate all MBQCs in a signed graph, which is further optimized with several carefully-crafted techniques to prune unpromising search spaces and enhance enumeration efficiency. Extensive experiments on real-world datasets demonstrate the efficiency, scalability, and effectiveness of our MBQC model and algorithms.

Fake news detection is a hot topic in the social media mining research community. Recent studies have shown that sentiment signals could significantly benefit the detection performance. However, most existing methods treat sentiment merely as auxiliary features, while the more sophisticated social sentiment interactions were rarely explored. In this paper, we propose a novel framework named ReFEND, which leverages the sentiment resonances among the social users (i.e., social sentiment resonances) and the sentiment relationship between news content and user comments to improve the detection performance. Specifically, we first utilize sentiment scorers to assess the sentiment of comments and identify users' emotional tendencies. Then we creatively construct a sentiment-aware multi-relational graph to capture social sentiment resonances evoked by the content and the interactions between comments and news. Next, we leverage the relational graph convolutional network (RGCN), which specializes in handling multi-relational graph data, to learn the interactions on sentiment-aware graph. To our best knowledge, this is the first effort to leverage social sentiment resonances for fake news detection. Experimental results on three datasets indicate that ReFEND significantly outperforms the state-of-the-art sentiment-based methods in terms of F1 and accuracy. Besides, ablation studies demonstrate the effectiveness of components designed in ReFEND.

Offloading a plethora of end workflows to edge servers in mobile edge computing (MEC) systems involves a series of coupled decision-making steps, includinghow muchedge resources will be allocated for each workflow,whichsubtasks will be offloaded, andhowto determine edge-end transaction prices to ensure system stability. Particularly, these decisions must jointly account for workflow characteristics, the resources available at each edge server, as well as local constraints (e.g., communication distance, task latency, and bandwidth conditions), which again increases the difficulty of optimizing the problem. However, no existing study addresses such joint optimization problems for these tightly coupled decisions. To fill this gap, a minimum-delay workflow partitioning algorithm is first designed to determine the optimal task offloading solution under various resource conditions. Based on this algorithm, two locality-based social welfare maximization models (basic and dynamic) are constructed. Specifically, for basic model, a multi-stage task matching game with the second lowest cost strategy is developed to determine the resource selection and pricing. For the dynamic model with uncertain requests, an online learning algorithm is introduced to track the dynamic valuations of mobile devices and to ensure that the resulting task allocation solution achieves an upper-bounded regret. Strict theoretical analysis demonstrates that our mechanism guarantees individual rationality, Nash Equilibrium, and stable approximation ratio. Simulation results verify the effectiveness and efficiency of our mechanism, and show that the proposed mechanisms obtain at most 18% higher social welfare than existing studies.

Concept reduction that preserves binary relations is an emerging reduction theory in the field of Formal Concept Analysis. Its core lies in reducing the number of concepts while ensuring that the original information is not lost, thereby significantly improving the efficiency of data processing. Based on Object Pictorial Diagram (OPD), this paper proposes a novel attribute-oriented concept reduction method that preserves complementary binary relations. First, this paper clarifies the definition of attribute-oriented concept reduction and presents a specific method for addressing it from the perspective of OPD. Against the backdrop of smart education's growing emphasis on data-driven decision-making, accurately diagnosing learners' knowledge states has become a core requirement for instructional reform and personalized tutoring. Practically, by integrating learners' response data to exercises, cognitive diagnosis is conducted by using the obtained attribute-oriented concept reduction results, enabling an in-depth analysis of learners' knowledge states and cognitive structures. Experimental results demonstrate that the proposed method exhibits high efficiency in both solving attribute-oriented concept reduction and performing cognitive diagnosis. The proposed method provides robust support for assessing learners' learning states and enhances the interpretability of various personalized learning applications.

Many Next-Generation consumer electronic devices would be distributed hybrid electronic systems, such as UAVs (Unmanned Aerial Vehicles) and smart electronic cars. The safety and risk control are the key issues for the sustainability of such consumer electronic systems. The modeling of hybrid electronic systems is difficult to be abstracted by traditional Petri Nets. This also makes the reachable marking graph unable to be applied to Petri Nets of the hybrid electronic systems. This paper proposes a novel Petri Net to model and analyze the hybrid electronic systems. We name it a Semi-continuous Colored Petri Net (SCPN) that inherits the excellent modeling capabilities and analysis methods of Petri Nets, and can formally depict hybrid quantities. In addition, we propose the construction algorithm for an SCPN reachable marking graph and prove its finiteness. Finally, we model and analyze an Adaptive Cruise Control (ACC) system of smart electronic cars as an example to prove the validity of SCPN. We use the proposed SCPN to model and analyze the running process of an ACC system under the continuous deceleration scenario of the front vehicle. The application study shows that the ACC system has logic flaws under the constant headway strategy when the front vehicle continues to decelerate. Based on this analysis, improvements to the SCPN of the ACC system are made, effectively enhancing its safety and logical correctness.

Aiming to improve the efficiency of urban traffic management, previous studies have achieved considerable traffic prediction accuracy. For example, methods based on time series analysis perform well in short-term traffic prediction, and neural networks show strong capabilities in processing complex nonlinear relationships within traffic data. However, previous studies also have the following two limitations: 1) a large amount of complex traffic data will increase the complexity of the model during training and further reduce the accuracy of the training results; 2) the large-scale distribution of traffic data leads to incomplete model training and data security issues. To address these issues, we propose a Formal Concept-enhanced Federated Graph Convolutional Network (FC-FedGCN), which adopts formal concept analysis to fully mine graph data and improve the training accuracy of the GCNs model. Under federated learning, the GCNs model can be trained independently on different clients, and the local model is optimized by sharing model parameters. Coupled with the premise of protecting data privacy, the integrity of the data is guaranteed and the training accuracy of the GCNs model is improved. We compare our model with various baseline models based on the PEMS datasets, and the results demonstrate that FC-FedGCN has significant advantages in traffic prediction, outperforming the comparison methods in multiple indicators.

Predicting information popularity in social networks has become a central focus of network analysis. While recent advancements have been made, most existing approaches rely solely on the final cascade size as the primary supervision signal for model optimization. This narrow focus limits the model generalization ability, particularly when faced with highly heterogeneous cascades. Additionally, in real-world scenarios, obtaining detailed social relationships is challenging, complicating effective structural feature learning. To address these issues, this paper proposes a semi-supervised model called Dual Variational Cascade AutoEncoders (DVCAE), which leverages parallel structural and temporal variational autoencoders for enhanced feature learning and popularity prediction. The model first aggregates multiple cascades into a global interaction graph, enabling structural information sharing across cascades. Then, it applies sparse matrix factorization-based graph embedding and graph filtering techniques on global and local cascade graphs respectively, generating initial node embeddings that are insensitive to topological perturbations. After that, two parallel variational autoencoders are designed to generate hidden representations for structural and temporal features respectively, with two self-supervised reconstruction losses integrated into the prediction loss to enrich supervision signals. Extensive experiments conducted on three real-world datasets demonstrate that DVCAE outperforms state-of-the-art models in terms of prediction accuracy.

Graph neural network (GNN) is a state-of-the-art technique for learning structural information from graph data. However, training GNNs on large-scale graphs is very challenging due to the size of real-world graphs and the message-passing architecture of GNNs. One promising approach for scaling GNNs is distributed training across multiple accelerators, where each accelerator holds a partitioned subgraph that fits in memory to train the model in parallel. Existing distributed GNN training methods require frequent and prohibitive embedding exchanges between partitions, leading to substantial communication overhead and limited the training efficiency. To address this challenge, we propose XDGNN, a novel distributed GNN training method that eliminates the forward communication bottleneck and thus accelerates training. Specifically, we design an explanation-guided subgraph expansion technique that incorporates important structures identified by eXplanation AI (XAI) methods into local partitions, mitigating information loss caused by graph partitioning. Then, XDGNN conducts communication-free distributed training on these self-contained partitions through training the model in parallel without communicating node embeddings in the forward phase. Extensive experiments demonstrate that XDGNN significantly improves training efficiency while maintaining the model accuracy compared with current distributed GNN training methods.

Wireless-powered multi-access edge computing (WP-MEC), as a promising computing paradigm with the great potential for breaking through the power limitations of wireless devices, is facing the challenges of reliable task offloading and charging power allocation. Towards this end, we formulate a joint optimization problem of wireless charging and computation offloading in socially-aware D2D-assisted WP-MEC to maximize the utility, characterized by wireless devices’ residual energy and the strength of social relationship. To address this problem, we propose a deep reinforcement learning (DRL)-based approach with hybrid actor-critic networks including three actor networks and one critic network as well as with Proximal Policy Optimization (PPO) updating policy. Further, to prevent the policy collapse, we adopt the PPO-clip algorithm which limits the update steps to enhance the stability of algorithm. The experimental results show that the proposed algorithm can achieved superior convergence performance and, meanwhile, improves the average utility efficiently compared to other baseline approaches.

The progress of smart cities is closely dependent on data, that are sensed by a number of mobile devices, and provide various intelligence services for the cities. However, with the increasing importance of data, privacy threats are increasing. Malicious actors can exploit eavesdropping and data analysis to illicitly access users’ sensitive information, which may encompass location data, preferences, social behavior, and smart home data. To this end, this article proposes a new certificateless blind signature (CLBS) scheme combined with a conditional revocation function to enhance user privacy protection in the smart city environment. Concretely, the scheme relies on an anonymous authentication mechanism to protect user identity privacy and uses an anonymous revocation mechanism to resist malicious data submission. In addition, the proposed scheme addresses the key escrow problem commonly encountered in traditional cryptosystems. The security of the scheme is verified by the random oracle model, and the bilinear pairing operation is avoided in the design to improve efficiency. Performance analysis results show that the proposed scheme improves the computational efficiency by at least 20% compared with the previous schemes.

Blended learning networks (BLNs) based on the integration of online learning networks and offline learning environments provide new opportunities and platforms for people to acquire and update useful knowledge and carry out all kinds of learning activities anytime and anywhere. Effective modeling and regulation of the knowledge dissemination process can accurately grasp its dissemination process, promote knowledge innovation and collaborative sharing among learners, and accelerate the maximization of knowledge dissemination. However, it is a challenge to establish a comprehensive dynamics model and adopt the optimal regulation for the knowledge dissemination process under the constraints of a limited budget in large-scale BLNs with diverse learners. To this end, we first explore the evolution process of knowledge dissemination in BLNs and the blended learning interaction process of learners. Based on the system dynamics modeling theory, a dynamics model of knowledge dissemination is established. Second, two kinds of effective regulation strategies are proposed. We establish an optimal regulation system intending to maximize the dissemination of knowledge and use the optimal control theory to tackle the optimal solution distribution of regulation strategies. Then, we propose a knowledge dissemination regulation task allocation method based on the collaborative participation of users, and the reverse auction theory is used to quickly solve the task allocation scheme while ensuring performance. Finally, we demonstrate the effectiveness of proposed models and methods through extensive simulation experiments based on real datasets.

Attributed social networks are pervasive in real life and play a crucial role in shaping various aspects of society. These networks not only capture the connections between individuals but also encompass the associated attributes and characteristics. Analyzing and understanding these attributes provide insights into social behaviors, information diffusion patterns, and the formation of influential communities. Consequently, we propose a novel algorithm for detecting fairness-aware maximal cliques in the attributed social networks. We extract the concept lattice of attributed social networks and quantify these concepts using the concept stability and fairness measures defined in this article. By utilizing the proposed fairness-aware distance, we identify fairness-aware maximal cliques within attributed social networks. The effectiveness of the algorithm is then validated using five real-world network datasets. Experimental results fully demonstrate the effectiveness and scalability of our approach in identifying key structures, analyzing attribute networks, and promoting the development of responsible computational systems.

The continuous interaction of users and information aggregation has become a social phenomena over massive social media platforms. However, the uncertainty of users’ behavior is leading great challenges to social networks analysis in terms of system structure, evolution characteristics, dynamic behavior, and so forth. Thus, this article proposes a formal user behavior modeling and analysis approach. First, aiming at identifying the behavior patterns of user activity sequence, we present a user activity transition system model based on stochastic Petri net (SPN), which can formally depict the process and structures of social users click activities. Then, the average number of tokens in each place, the probability density function of the tokens, the token flow rate of transitions, and the time spent in each state are analyzed by isomorphic it into a Markov chain (MC), respectively. These four indicators are used to evaluate the performance of the proposed system model. The experimental results demonstrate that the proposed approach can help us to understand the rules of users’ first activity and activity preferences, so as to provide practical suggestions for the development of social networking platforms and content recommendation.

The capacity to create “fake” videos has recently raised concerns about the reliability of multimedia content. Identifying between true and false information is a critical step toward resolving this problem. On this issue, several algorithms utilizing deep learning and facial landmarks have yielded intriguing results. Facial landmarks are traits that are solely tied to the subject’s head posture. Based on this observation, we study how Head Pose Estimation (HPE) patterns may be utilized to detect deepfakes in this work. The HPE patterns studied are based on FSA-Net, SynergyNet, and WSM, which are among the most performant approaches on the state-of-the-art. Finally, using a machine learning technique based on K-Nearest Neighbor and Dynamic Time Warping, their temporal patterns are categorized as authentic or false. We also offer a set of experiments for examining the feasibility of using deep learning techniques on such patterns. The findings reveal that the ability to recognize a deepfake video utilizing an HPE pattern is dependent on the HPE methodology. On the contrary, performance is less dependent on the performance of the utilized HPE technique. Experiments are carried out on the FaceForensics++ dataset that presents both identity swap and expression swap examples. The findings show that FSA-Net is an effective feature extraction method for determining whether a pattern belongs to a deepfake or not. The approach is also robust in comparison to deepfake videos created using various methods or for different goals. In the mean the method obtain 86% of accuracy on the identity swap task and 86.5% of accuracy on the expression swap. These findings offer up various possibilities and future directions for solving the deepfake detection problem using specialized HPE approaches, which are also known to be fast and reliable.

Synchronization plays a crucial role in computer systems, providing support for system security, data consistency, and coordination. It contributes to the establishment and application of trust, security, and dependability in distributed systems and concurrent computing to a significant extent. This article makes innovative contributions in the field of synchronic distance in Petri net. We provide refined definitions for the hierarchical classification of synchronic levels in Petri net, proposing the concepts of absolute synchronization, strong synchronization, and extended synchronization based on different conditions. Furthermore, we propose an innovative method for calculating synchronic distance. This method can automate the calculation of synchronic distance between any two transitions using computer computation, resulting in improved accuracy and reduced errors. This novel approach provides an effective tool for system security and trustworthy modeling, as accurate synchronic distance calculations allow for better evaluation of synchronic distance between different transitions, leading to the identification of potential security vulnerabilities and design flaws, thereby enhancing the credibility of decision-making and promoting the reliability of models and analysis results. To validate the proposed method, we introduce a specific example of a Petri net with concurrency, demonstrate the practicality and effectiveness of the proposed method and algorithm through analysis of this example. Our work extends the research on Petri net synchronic distance, further advancing the understanding and exploration of this field.

Emotion estimation from face expression analysis is nowadays a widely-explored computer vision task. In turn, the classification of expressions relies on relevant facial features and their dynamics. Despite the promising accuracy results achieved in controlled and favorable conditions, the processing of faces acquired at a distance, entailing low-quality images, still suffers from a significant performance decrease. In particular, most approaches and related computational models become extremely unstable in the case of the very small amount of useful pixels that is typical in these conditions. Therefore, their behavior should be investigated more carefully. On the other hand, real-time emotion recognition at a distance may play a critical role in smart video surveillance, especially when controlling particular kinds of events, e.g., political meetings, to try to prevent adverse actions. This work compares facial expression recognition at a distance by: 1) a deep learning architecture based on state-of-the-art (SOTA) proposals, which exploits the whole images to autonomously learn the relevant embeddings; 2) a machine learning approach that relies on hand-crafted features, namely the facial landmarks preliminarily extracted using the popular Mediapipe framework. Instead of using either the complete sequence of frames or only the final still image of the expression, like current SOTA approaches, the two proposed methods are designed to use rich temporal information to identify three different stages of emotion. Expressions are time-split accordingly into four phases to better exploit their temporal-dependent dynamics. Experiments were conducted on the popular Extended Cohn-Kanade dataset (CK+). It was chosen for its wide use in related literature, and because it includes videos of facial expressions and not only still images. The results show that the approach relying on machine learning via hand-crafted features is more suitable for classifying the initial phases of the expression and does not decay in terms of accuracy when images are at a distance (only 0.08% of decay). On the contrary, deep learning not only has difficulties classifying the initial phases of the expressions but also suffers from relevant performance decay when considering images at a distance (52.68% accuracy decay).

Cohesive subgraph mining on attributed social networks is attracting much attention in the realm of graph mining and analysis. Most existing studies on cohesive subgraph mining over attributed social networks neglect the fairness of attributes, which lead to difficulties in deploying responsible applications. Toward this end, this article formulates a new problem by introducing fairness into cliques model to mine the absolute fair cliques from attributed social networks. Specifically, this article adopts formal concept analysis (FCA) methodology to represent the given attributed social network, and extracts a set of special attributed equiconcepts to further return the absolute fair maximal cliques. Then, we develop an efficient absolute fair cliques detection algorithm AFCMiner for the cases of single-dimensional attributed social networks, multivalued attributed social networks, as well as multidimensional attributed social networks. Extensive experiments are conducted for demonstrating that the proposed AFCMiner algorithm can significantly reduce the time for finding absolute fair cliques with the correctness guarantee. Finally, a case study is also presented for uncovering the usefulness of our model.

Thanks to the booming development of artificial intelligence, 5G technology, and intelligent manufacturing technology, numerous intelligent edge devices contained in the industrial Internet of Things (IIoT) are endowed with the ability to mine knowledge from perceived massive data. Knowledge-driven IIoT plays an unprecedented role in application fields such as cyber-physical systems and Industry 4.0. However, knowledge is generally scattered across the distributed edge devices of IIoT. Therefore, in order to further achieve the edge intelligence in IIoT, it is very important to explore an efficient edge knowledge sharing method. In this article, we establish a decentralized knowledge sharing platform in IIoT. First, for public knowledge, a dynamics model that can quantitatively describe its sharing process is established by using the system dynamics theory. Furthermore, a control method for maximizing public knowledge sharing under constraints based on the optimal control theory is presented. Second, for private knowledge, a trusted transaction control method based on blockchain technology is proposed. By developing both smart contract and lightweight consensus mechanism, the efficient peer-to-peer sharing of private knowledge is realized, and the integrity of knowledge and the privacy of participants are protected. The results of extensive experiments show that the proposed method can eliminate the obstacles of knowledge sharing among edge devices in IIoT, and further promote the development of edge intelligence empowered 5G-enabled IIoT applications.

Massive Open Online Courses (MOOCs) belong to a new cloud-based service in education that suffers from low completion rates. Effective pre-learning intervention services, such as recommending courses with a high probability of completion or filtering courses with a very low probability of completion, will encourage students to spend more time and energy on proper courses, thus can reduce the dropout ratio. In practice, intervention services are introduced when students are predicted to drop out. However, existing methods concentrate on analyzing students’ learning actions and predicting final dropout after a period of enrollment, which are insufficient in preventing students from enrolling in unsuitable courses and withdrawing mid-way. This paper presents a neural network-based Explainable Self-supervised Model (ESM) to predict MOOC dropout before enrollment. Specifically, the student's learning actions on an unenrolled course are estimated using previous logs by the neural network. And then, the action's contribution to the completion of a course is calculated in a similar way. Therefore, the probability of completion for an unenrolled course is predicted by aggregating the learning actions and their contribution to the completion. To train the neural network, a self-supervised training strategy is proposed, where enrolled courses in the training data are randomly selected as validation in each epoch. The ESM outperforms existing methods in terms of prediction accuracy and efficiency. The average increment of Area Under the ROC Curve (AUC) and F-score (F1) in the two MOOCs datasets, XuetangX and KDDCUP, are 8.3% and 0.6%, respectively. Furthermore, the two pre-learning intervention services named courses recommendation and courses filtration are proposed. When courses are recommended, the completion rate increased from 22% to 60% in XuetangX, and from 27% to 45% in KDDCUP. By filtering courses predicted with low completion probability, 40% wasted time in uncompleted courses will be saved in XuetangX.

With the advent of 5G, Mobile Edge Computing (MEC), a promising computing paradigm sits closer to users than cloud computing, is being broadly used in various Internet of Things (IoT) applications, and achieve high-quality user experience. Task offloading, as a critical research issue in MEC, is playing an important role in optimizing computational resources and management. However, many tasks are executed dependent on the computational results of other tasks. Moreover, in the case of offloading tasks with other devices, it is often required to consider the success rate of offloading, since not all users are willing to lend their mobile devices to others for task execution. To address this challenge, by taking social relationships between users into account, this paper intends to combine computational resources of local devices and edge clouds and provide more flexible offloading and execution solutions, for achieving the efficient offloading of dependent tasks with the joint consideration of network latency and energy consumption. This paper develops a dependent task offloading strategy based on Bipartite Graph Matching. Extensive simulations are conducted for validating the effectiveness of our proposed strategy. Experimental results demonstrate that our proposed strategy can significantly minimize the overhead compared with other baseline strategies. In particular, the overhead is reduced 8.2%, compared with the strategy which consider the Device-to-Device (D2D) offloading only.

Identifying the optimal groups of users that are closely connected and satisfy some ranking criteria from an attributed social network attracts significant attention from both academia and industry. Skyline query processing, a multicriteria decision-making optimized technique, is recently embedded into cohesive subgraphs mining in graphs/social networks. However, the existing studies cannot capture the fuzzy property of connections between users in social networks. To fill this gap, in this article, we formulate a novel model of the skyline$(\lambda,k)$-cliques over a fuzzy attributed social network and develop a formal concept analysis (FCA)-based skyline$(\lambda,k)$-cliques identification algorithm. Specifically,$\lambda $can be regarded as a quality control parameter for measuring the stability of the cohesive groups. Extensive experimental results conducted on three real-world datasets demonstrate the effectiveness of the skyline$(\lambda,k)$-clique model in a fuzzy attributed social network. Furthermore, an illustrative example is executed for revealing the usefulness of our model. It is expected that our proposed skyline$(\lambda,k)$-clique model can be widely used in various graph-based computational social systems, such as optimal team formation in crowdsourcing, and group recommendation in social networks.

The wide availability of online social networks (OSNs) facilitates positive information spread and sharing. However, the high autonomy and openness of the OSNs also allow for the rapid spread of negative information, such as unsubstantiated rumors and other forms of misinformation that often elicit widespread public cognitive misleads and huge economic losses. Therefore, how to effectively control the negative information spread accompanied by positive information has emerged as a challenging issue. Unfortunately, this issue still remains largely unexplored to date. To fill this gap, we propose an efficient feedback control mechanism for the simultaneous spread of the positive and negative information in OSNs. Specifically, a novel computational model is first proposed to present the temporal dynamics of the positive and negative information spread. Furthermore, the proposed mechanism restrains the negative information spread with minimal system expenses by devising and performing three synergetic intervention strategies. Technically, this mechanism intensively evaluates the number of seed users performing three intervention strategies. Besides, each seed user performs the received control task independently, and then the control plan for the next time step is adjusted dynamically according to the previous feedback results. Finally, we evaluate the efficiency of the proposed mechanism based on the extensive experimental results obtained from two real-world networks.

Interconnectivity and smart automation of Internet of Things in recent times have led to the concept of Industry 4.0. Together with the improvement in productivity and new business models, employment conditions should take advantage of these new technologies. Safety in the workplace is one of the most sensitive topics on matters that needs targeted and accurate solutions. The safety can be guaranteed by investigating the attention states of the workers, and in particular, their drowsiness levels. Several technologies have faced this problem by using biometrics, but how many of them are applicable in a real-case-use scenario of Industry 4.0? This article aims to answer this question by discussing available data and methods that can be used in specific workplaces. We highlight their limitations and accuracy to sketch out the recent literature that may contribute to worker safety in Industry 4.0. Finally, we point out a gap that needs to be filled in order to implement these strategies on a large scale.

Knowledge graph describes entities by numerous RDF data (subject-predicate-object triples), which has been widely applied in various fields, such as artificial intelligence, Semantic Web, entity summarization. With time elapses, the continuously increasing RDF descriptions of entity lead to information overload and further cause people confused. With this backdrop, automatic entity summarization has received much attention in recent years, aiming to select the most concise and most typical facts that depict an entity in brief from lengthy RDF data. As new descriptions of entity are continually coming, creating a compact summary of entity quickly from a lengthy knowledge graph is challenging. To address this problem, this article first formulates the problem and proposes a novel approach of Incremental Entity Summarization by leveraging Formal Concept Analysis (FCA), called IES-FCA. Additionally, we not only prove the rationality of our suggested method mathematically, but also carry out extensive experiments using two real-world datasets. The experimental results demonstrate that the proposed method IES-FCA can save about 8.7 percent of time consumption for all entities than the non-incremental entity summarization approach KAFCA at best. As for the effectiveness, IES-FCA outperforms the state-of-the-art algorithms in terms of$F1-measure$,$MAP$, and$NDCG$.

The increasing of group polarization on social media seriously impacts on the health of public discourse and information dissemination. At present, detecting polarized structures in signed networks is well-motivated for studying the group polarization on social media. However, most studies restricted the number of polarized structures to only two, while neglecting the real-world scenario where signed networks consist of multiple polarized structures, that is an unreasonable assumption. To conquer the limitations of the existing work, in this paper, we present a novel cohesive subgraph model based on structural clusterable theory, named maximal multipolarized clique (MMC), which can be partitioned into k polarized subcliques such that the edges in subcliques are positive and the edges between subcliques are negative. This paper formulates the problem of Maximal Multipolarized Cliques Search (MMCS) in signed networks which is proved to be NP-hard. To address this problem, we first devise powerful pruning rules to reduce the signed network significantly and further develop an efficient algorithm to search all maximal multipolarized cliques in the reduced signed network. The experimental results on real-world signed networks demonstrate the efficiency and effectiveness of our algorithm.

With the rapid development of information technology, the huge amount of information is booming exponentially, and it becomes a key challenge to find the information users required from the massive information. To tackle this challenge, this paper focuses on developing a novel personalized recommendation approach based on complementary context-enhanced concept lattice. To be specific, the proposed approach constructs the concept lattices for both formal context and complementary context of user-item interactive data, and then adopts a Formal Concept Analysis based association rule recommendation algorithm for obtaining two recommendation result sets separately, and ultimately analyzes the acquisition of the final resulting recommendation for each of these two recommendation result sets under different situations. Taking movie recommendation as an illustrative example, more accurate user-preferred movie recommendation can be achieved. Compared with the traditional personalized recommendation algorithms, it is confirmed that our recommendation approach is more reasonable and effective in the real recommendation systems.

Mobile social networks (MSNs) provide real-time information services to individuals in social communities through mobile devices. However, due to their high openness and autonomy, MSNs have been suffering from rampant rumors, fraudulent activities, and other types of misuses. To mitigate such threats, it is urgent to control the spread of fraud information. The research challenge is: how to design control strategies to efficiently utilize limited resources and meanwhile minimize individuals' losses caused by fraud information? To this end, we model the fraud information control issue as an optimal control problem, in which the control resources consumption for implementing control strategies and the losses of individuals are jointly taken as a constraint called total cost, and the minimum total cost becomes the objective function. Based on the optimal control theory, we devise the optimal dynamic allocation of control strategies. Besides, a dynamics model for fraud information diffusion is established by considering the uncertain mental state of individuals, we investigate the trend of fraud information diffusion and the stability of the dynamics model. Our simulation study shows that the proposed optimal control strategies can effectively inhibit the diffusion of fraud information while incurring the smallest total cost. Compared with other control strategies, the control effect of the proposed optimal control strategies is about 10% higher.

One of the most important sources of revenue for social media platforms, like, Twitter, Facebook, Reddit, etc., is advertising. An effective social media advertising plan moves people from awareness and interest in desire and action. Despite the potentiality, campaigns and marketing strategies should be improved. One of the challenges is to identify the right target audience at the right time, considering both communities of interests and locations and the development of these conditions along the timeline. This is crucial to create the right communication strategy and the right advertising message. This paper proposes a context-aware ad-targeting methodology using time, locations, and inferring users' interests by analyzing published content. The method relies on a fuzzy extension of Triadic Formal Concept Analysis for identifying Location-based and Content-based communities of users. Then, a task of community fusion takes place, named Join, for matching a target audience. The matching may be tuned for identifying a wide or narrow community and implementing a fine-grained ad targeting. Experimental results are given.

The dramatic spread of online social network services, such as Facebook, Twitter, Instagram, and Google+, has led to increasing awareness of the power of incorporating social elements into a variety of data-centric applications. These applications, in recent years, apply various sensors with social media platforms to continuously collect massive data that can be directly associated with human interactions. This phenomenon has led to the creation of numerous social sensing systems, such as Biketastic, BikeNet, CarTel, and Pier, which use social sensors (i.e., users) for a variety of social sensing systems and applications. Social sensing has become an emerging and promising sensing paradigm that relies on the voluntary cooperation of users equipped with embedded or integrated sensors.

Mobile Edge Computing (MEC) is providing cloud computing capabilities within the radio access networks and offering a new paradigm to liberate the mobile devices from heavy computational workloads. Importantly, MEC can effectively reduce latency, avoid congestion, and prolong the battery lifetime of mobile devices by offloading the computation tasks from the mobile devices to a physically proximal MEC servers. Particularly, Virtual Machines (VMs) scheduling is a critical issue for tasks offloading and computation in MEC. Regarding to the VMs scheduling problem in MEC environmnet, this paper pioneers the use of Formal Concept Analysis (FCA) methodology for identifying the mapping from tasks to VMs. Specifically, the VMs profile and tasks descriptions are initially characterized as the formal contexts, respectively. With the constructed formal contexts, the corresponding formal concepts which refer to the rules set, are then generated. To better infuse the rules set of VMs and tasks, this paper defines a similarity measurement between formal concepts of VMs and tasks. Consequently, the matching problem from a given task to a virtual machine is to return the expected virtual machine according to the principle of maximum similarity degree between formal concepts of virtual machine and task. Extensive simulations are conducted with a real dataset for the validation of feasibility and effectiveness of the proposed approach. Specifically, the proposed approach can significantly reduce the energy consumption around 28 percent comparing to the approach without consideration of energy consumption. Overall, It is demonstrated that FCA-based VMs scheduling is a novel solution for a sustainable VMs scheduling in MEC environment.

Mobile edge computing (MEC) is providing a promising solution for augmenting the computing and storage capacity of mobile devices by exploiting the available resources at the network edge. Among the various Internet of Things (IoT) applications, MEC could help us to narrow the gap between the requirements of IoT applications and the limited resources of IoT devices and to achieve the energy-efficient communication and computing. Importantly, the upper and edge infrastructure of cloud computing should effectively collaborate for executing the complex tasks which are requested by mobile users. In particular, community cloud computing, as a novel computational model for a specific community with common concerns (such as security, compliance, and jurisdiction), can make full use of the spare resources of networked computers to provide the facilities so that the community gains services from them. However, how to allocate the subtasks into community clouds and edge community clouds (cloudlets) is becoming a critical challenge. To tackle this challenge, this paper first proposes a two-layer multi-community-cloud/cloudlet social collaborative paradigm, called 2L-MC3for MEC. Further, we formulate a problem on tasks allocation in community clouds/cloudlets by jointly taking task offloading, tasks and clouds profiles into account. To address this problem, we devise a bi-level programming model for tasks allocation. Extensive simulations are conducted for demonstrating that the proposed approach can achieve the relative global performance for satisfying the each metric comparing to the other approaches.

Depression is becoming a serious global health problem worldwide, with an increasing number of patients suffering from anxiety and other disorders. Our work aims to provide the appropriate social support (SS) to the prevention of depression for highly anxious undergraduates. We used 1425 undergraduates from 18 universities in China via a cluster random sampling method for the survey on the self-rating anxiety scale, the self-rating depression scale, and the SS scale for anxiety and depression. Based on the collected questionnaire data, we first reveal that the distribution of both anxiety data and depression data follows a Gaussian distribution. Then, a Gaussian mixture model is adopted for clustering these data in terms of anxiety index and depression index. According to the observations extracted from the clusters, the correlation among anxiety, depression, and SS is investigated by a correlation analysis method. Finally, the corresponding moderating effect of SS between anxiety and depression is figured out via the hierarchical multiple regression analysis. The detailed analysis indicates that the high-level SS, such as the help and support from individual's friends or family members, could reduce the risk for depression from highly anxious undergraduates.

The increasing popularization of large-scale online social networks (OSNs) facilitates information sharing. Users are able to diffuse positive and negative information independently owing to the high openness of the OSNs. Due to the intervention of user’s emotions and social relationships, the positive and negative information diffusion exhibits a complicated dynamic coupling diffusion process, in which the negative information diffusion can cause social panic and confusion. However, prior works mainly focus on the diffusion of single type of information. To fill this gap, this paper aims to investigate the dynamic diffusion process of the positive and negative information and control the negative information diffusion timely. Specifically, we first establish a coupling diffusion model to characterize the dynamic diffusion process under the coexistence of positive and negative information, then derive the critical condition for the negative information diffusion and certify the stability of the diffusion model. Furthermore, we propose two collaborative control strategies to persuade and guide users to diffuse the positive information simultaneously. Then, the issue of minimizing the total system costs is transformed to an optimal control problem. Finally, we prove the existence and uniqueness of optimal control solutions and obtain the dynamic distribution of optimal control strategies over time to minimize system costs. The experimental results obtained from two real-world datasets verify the effectiveness of our model and the high efficiency of the collaborative control strategies.

Single image super resolution has achieved a significant breakthrough with the development of deep learning technology. Among these approaches based on deep learning, the mainstream method is to build a cascading network and attempt to add more learning layers. However, as the depth of the model increases, features far away from the reconstruction layer are less considered in the reconstruction process. In this paper, we propose a novel model based on a refined densely connected network for super-resolution reconstruction tasks. By utilizing densely connected paths in the model, we can significantly shorten the distance between the feature maps from different levels and the reconstruction layer. Besides, an inception-like structure is employed to replace the ordinary convolutional layer to take full advantage of the contextual information. Moreover, quantities of 1 ×1 filters are used to ensure an acceptable model size. Extensive experiments are conducted for demonstrating that the proposed method can achieve the state-of-the-art performance with smaller model size.

With the exponential growth of data volume, big data have placed an unprecedented burden on current computing infrastructure. Dimensionality reduction of big data attracts a great deal of attention in recent years as an efficient method to extract the core data which is smaller to store and faster to process. This paper aims at addressing the three fundamental problems closely related to distributed dimensionality reduction of big data, i.e., big data fusion, dimensionality reduction algorithm and construction of distributed computing platform. A chunk tensor method is presented to fuse the unstructured, semi-structured and structured data as a unified model in which all characteristics of the heterogeneous data are appropriately arranged along the tensor orders. A Lanczos based high order singular value decomposition algorithm is proposed to reduce dimensionality of the unified model. Theoretical analyses of the algorithm are provided in terms of storage scheme, convergence property and computation cost. To execute the dimensionality reduction task, this paper employs the transparent computing paradigm to construct a distributed computing platform as well as utilizes a four-objectives optimization model to schedule the tasks. Experimental results demonstrate that the proposed holistic approach is efficient for distributed dimensionality reduction of big data.

Participatory sensing is a promising sensing paradigm that enables collection, processing, dissemination and analysis of the phenomena of interest by ordinary citizens through their handheld sensing devices. Participatory sensing has huge potential in many applications, such as smart transportation and air quality monitoring. However, participants may submit low-quality, misleading, inaccurate, or even malicious data if a participatory sensing campaign is not launched effectively. Therefore, it has become a significant issue to establish an efficient participatory sensing campaign for improving the data quality. This article proposes a novel five-tier framework of participatory sensing and addresses several technical challenges in this proposed framework including: (1) optimized deployment of data collection points (DC-points); and (2) efficient recruitment strategy of participants. Toward this end, the deployment of DC-points is formulated as an optimization problem with maximum utilization of sensor and then a Wise-Dynamic DC-points Deployment (WD3) algorithm is designed for high-quality sensing. Furthermore, to guarantee the reliable sensing data collection and communication, a trajectory-based strategy for participant recruitment is proposed to enable campaign organizers to identify well-suited participants for data sensing based on a joint consideration of temporal availability, trust, and energy. Extensive experiments and performance analysis of the proposed framework and associated algorithms are conducted. The results demonstrate that the proposed algorithm can achieve a good sensing coverage with a smaller number of DC-points, and the participants that are termed as social sensors are easily selected, to evaluate the feasibility and extensibility of the proposed recruitment strategies.

Social recommendation has been popular and successful in various urban sustainable applications such as online sharing, products recommendation and shopping services. These applications allow users to form several implicit social networks through their daily social interactions. The users in such social networks can rate some interesting items and give comments. The majority of the existing studies have investigated the rating prediction and recommendation of items based on user-item bipartite graph and user-user social graph, so called social recommendation. However, the spatial factor was not considered in their recommendation mechanisms. With the rapid development of the service of location-based social networks, the spatial information gradually affects the quality and correlation of rating and recommendation of items. This paper proposes spatial social union (SSU), an approach of similarity measurement between two users that integrates the interconnection among users, items and locations. The SSU-aware location-sensitive recommendation algorithm is then devised. We evaluate and compare the proposed approach with the existing rating prediction and item recommendation algorithms subject to a real-life data set. Experimental results show that the proposed SSU-aware recommendation algorithm is more effective in recommending items with the better consideration of user's preference and location.

Mobile social networks (MSNs) facilitate connections between mobile users and allow them to find other potential users who have similar interests through mobile devices, communicate with them, and benefit from their information. As MSNs are distributed public virtual social spaces, the available information may not be trustworthy to all. Therefore, mobile users are often at risk since they may not have any prior knowledge about others who are socially connected. To address this problem, trust inference plays a critical role for establishing social links between mobile users in MSNs. Taking into account the nonsemantical representation of trust between users of the existing trust models in social networks, this paper proposes a new fuzzy inference mechanism, namely MobiFuzzyTrust, for inferring trust semantically from one mobile user to another that may not be directly connected in the trust graph of MSNs. First, a mobile context including an intersection of prestige of users, location, time, and social context is constructed. Second, a mobile context aware trust model is devised to evaluate the trust value between two mobile users efficiently. Finally, the fuzzy linguistic technique is used to express the trust between two mobile users and enhance the human's understanding of trust. Real-world mobile dataset is adopted to evaluate the performance of the MobiFuzzyTrust inference mechanism. The experimental results demonstrate that MobiFuzzyTrust can efficiently infer trust with a high precision.

Community Cloud Computing is an emerging and promising computing model for a specific community with common concerns, such as security, compliance and jurisdiction. It utilizes the spare resources of networked computers to provide the facilities so that the community gains services from the cloud. The effective collaboration among the community clouds offers a powerful computing capacity for complex tasks containing the subtasks that need data exchange. Selecting the best group of community clouds that are the most economy-efficient, communication-efficient, secured, and trusted to accomplish a complex task is very challenging. To address this problem, we first formulate a computational model for multi-community-cloud collaboration, namely$MC^{3}$. The proposed model is then optimized from four aspects: minimizing the sum of access cost and monetary cost, maximizing the security-level agreement and trust among the community clouds. Furthermore, an efficient and comprehensive selection algorithm is devised to extract the best group of community clouds in$MC^{3}$. Finally, the extensive simulation experiments and performance analysis of the proposed algorithm are conducted. The results demonstrate that the proposed algorithm outperforms the minimal set coverings based algorithm and the random algorithm. Moreover, the proposed comprehensive community clouds selection algorithm can guarantee good global performance in terms of access cost, monetary cost, security level and trust between user and community clouds.

Micro-blog marketing has become a main business model for social networks nowadays. On social networking sites (e.g., Twitter), micro-blog marketing enables the advertisers to put ads to attract customers to buy their products. During this process, a rather key step for the success of advertisers is to conduct marketing researches to discover which micro-blog users are their potential customers who can greatly promote their products to other customers so that the advertising investment can be greatly reduced. This problem is considered as “influence maximization” issue. In this paper and in attempt to discover the influential users in micro-blog marketing, we try to analyze the influences of nodes in a micro-blog network and propose a Community Scale-Sensitive Maxdegree (CSSM) algorithm for maximizing the influences when placing ads. Experimental results on the very hot micro-blog service (i.e., Twitter dataset) demonstrate that our proposed CSSM algorithm significantly outperforms other related node selection strategies, in terms of the influence spread and time complexity.

The internal delay methods for sequential equivalent sampling are widely used in the digital storage oscilloscope and limited completely by maximum operating frequency of embedded system. A new external delay technology for equivalent sampling oscilloscope is presented in this paper. The technology is based on the external programmable delay chips, which provide much shorter delay time for equivalent sampling rate as well as much higher operating frequency. With real time sampling and equivalent time sampling, the embedded system could perform rapid and effective measurement for the periodic signal which is either fast-varying or slow-varying and also be able to take real time sampling to the single-shot signals.

Attribute reduction plays an important role in rough set theory. It is an important application in data mining. In this paper, we focus on discussing the relation between set covering and attribute reduction in rough set theory. Based on the equivalence between minimal set covering and minimal attribute reduction sets, attribute reduction graph (ARG) is constructed. A novel algorithm to find the minimal attribute reduction sets, which is based on combination of binary discernibility matrix and graph theory is proposed in this paper. This algorithm demonstrates its efficiency and feasibility by an example.

Time series data is a series of observation data accord- ing to a certain time sequence. It has been penetrate various field. This paper applies Rough set to the knowledge dis- covery of time series. The process of knowledge discovery in time series includes preprocessing of time series data, at- tributes selection and similarity sequence searching. Then, the time series is partitioned to a set of pattern(each pattern represents a trend of time series)by mobile window method. An information table is formed by the most important pre- dicting attributes and target attribute which in the trend- ing structure sequence identified from each pattern. This information table is suitable for the Rough set to discover knowledge. The extracted rules can predict the time series behavior in the future. We demonstrate our method on time series stock market data.