Search

Automatic rumor detection is critical for maintaining a healthy social media environment. The mainstream methods generally learn rich features from information cascades by modeling the cascade as a tree or graph structure where edges are built based on interactions between a tweet and retweets. Some psychology studies have empirically shown that users’ various subjective factors always cause the uncertainty of interactions such as differences among interactive behavior activation thresholds or semantic relevancy. However, previous works model interactions by employing a simple fully connected layer on fixed edge weights in the graph and cannot reasonably describe this inherent uncertainty of complex interactions. In this article, inspired by the fuzzy theory, we propose a novel neuro-fuzzy method, fuzzy graph convolutional networks (FGCNs), to sufficiently understand uncertain interactions in the information cascade in a fuzzy perspective. Specifically, a new strategy of graph construction is first designed to convert each information cascade into a heterogeneous graph structure with the consideration of explicit interactive behaviors between a tweet and its retweet, as well as implicit interactive behaviors among retweets, enriching more structural clues in the graph. Then, we improve graph convolutional networks by incorporating edge fuzzification (EF) modules. The EFs adapt edge weights according to predefined membership to enhance message passing in the graph. The proposed model can provide a stronger relational inductive bias for expressing uncertain interactions and capture more discriminative and robust structural features for rumor detection. Extensive experiments demonstrate the effectiveness and superiority of FGCN on both rumor detection and early rumor detection.

Resilient estimation/control networked systems against denial-of-service (DoS) attacks or false data injection (FDI) attacks have been extensively studied. However, most of the existing works focus on the deterministic networked systems. Because of the highly hybrid dynamic nature of the stochastic networked control systems (NCSs) under coordinated DoS and FDI attacks, it is a very challenging problem to consider FDI attack estimation and DoS-resilience control of the stochastic NCSs simultaneously. To tackle this issue, this article presents a piecewise extended observer-based control framework of stochastic NCSs subject to both DoS and FDI attacks. This simplifies the observer error system modeling of the attacked stochastic NCSs with immeasurable states in the presence of DoS attacks and FDI attacks, thus reducing the difficulty of observer-based resilient controller design. An attack-parameter-dependent time-varying stochastic Lyapunov function is introduced to analyze the mean square exponential stability of the resultant error system, and a codesign method of a piecewise extended observer gain and $L_{2}$ controller gain is proposed as well. Finally, simulation studies on an aircraft system are performed to verify the effectiveness of the proposed method.

This article investigates the problem of resilient dynamic event-triggered control for networked unmanned surface vehicles (USVs) with external disturbance in the presence of transmission delay and aperiodic DoS attacks. First, a network-based USV control system model is constructed and converted into a state space. Second, a resilient dynamic event-triggered scheme is designed to save system communication resources under aperiodic DoS attacks. Third, to ensure the security of the networked USV control system and to block communication with unknown jamming signals, a time-delay switching system model based on resilient event-triggered mechanism under aperiodic DoS attack is established. Then, by combining the Lyapunov method and linear matrix inequalities (LMIs), the criterion for asymptotic stability analysis and controller synthesis conditions of the resulting switching system is derived. Approach efficiency is validated by numerical simulation.

In this article, the stochastic analysis and $H_{\infty }$ controller design problems of networked systems with packet dropouts and false data injection attacks are investigated. Different from the existing literature, we focus on the linear networked systems with external disturbances and both sensor–controller channel and controller–actuator channel are studied. First, we present a discrete-time modeling framework that leads to a stochastic closed-loop system with randomly varying parameters. To facilitate the analysis and $H_{\infty }$ control of resulting discrete-time stochastic closed-loop system, an equivalent yet analyzable stochastic augmented model is further constructed by matrix exponential computation. Based on this model, a stability condition is derived in the form of linear matrix inequality (LMI) with the aid of a reduced-order confluent Vandermonde matrix, Kronecker product operation, and law of total expectation. Specifically, the dimension of the LMI obtained in this article does not increase as the upper bound of consecutive packet dropouts does, which is also different from the existing literature. Subsequently, a desired $H_{\infty }$ controller is obtained such that the original discrete-time stochastic closed-loop system is exponentially mean-square stable with a prescribed $H_{\infty }$ performance. Finally, a numerical example and a direct current motor system are exploited to substantiate the effectiveness and practicability of the designed strategy.