Accurate multivariate time series forecasting (MTSF) is critical for intelligent web services in Web of Things. When confronted with unseen multivariate time series (MTS), the industry typically invests significant time and resources in training multiple models to identify the optimal model for deployment. This paper proposes a novel, efficient, and scalable MTSF model selection method that directly selects suitable MTSF methods based on data characteristics without extensive model training. Model selection is a core component of AutoML, which has made significant progress in recent years. However, existing methods incur high operational costs and cannot be directly applied to MTSF tasks. Moreover, there is a lack of a comprehensive and cohesive public time series library for MTSF model selection. To address these challenges, we compile the first large heterogeneous labeled MTSF model selection dataset, called the ModelPile, which covers 41 mainstream datasets across 11 domains. We then propose AutoMTSF, a large model-enabled model selection method that transforms the MTSF model selection problem into a time series classification problem and utilizes the ModelPile to unlock large-scale multi-dataset training. AutoMTSF first uses the pre-trained large model to encode raw MTS. Given the coarse-grained limitations of large model encoding, Recursive Temporal Pattern Feature (RTPF) is proposed to capture both fine-grained and global temporal feature evolution, thereby effectively mapping data characteristics to the MTSF method space. Experiments comparing AutoMTSF with 2 baselines, 17 MTSF methods, and 4 large time series models show that AutoMTSF outperforms state-of-the-art methods while maintaining comparable execution time. This work represents a critical step in validating the accuracy and efficiency of large model-enabled classification for MTSF.

Subgraph Anomaly Detection (SAD) is crucial for identifying groups that deviate from the regular pattern within graphs, which benefits different domains such as financial fraud and network security. However, current studies rely on traditional node detection methods and fixed sampling strategies of subgraph structures, which makes it difficult to learn the pattern collaboration behavior of subgraphs. To address this limitation, this paper proposes a novel unsupervised framework named PC-SAD. The PC-SAD framework first employs an improved Graph AutoEncoder to identify core anomaly nodes by capturing multi-scale neighborhood information. Starting from these core anomaly nodes, we sample candidate subgraphs with path, tree, and cyclic structures, and enhance them according to the characteristics of the subgraph structures. Subsequently, candidate subgraphs are fed into the proposed Pattern Collaboration-based Graph Contrastive Learning method to generate collaborative pattern embeddings, thereby distinguishing anomaly subgraphs. The experimental results show that PC-SAD outperforms the state-of-the-art baseline methods on four benchmark datasets, which proves that PC-SAD is an effective solution to detect anomaly subgraphs.

Recently, the research on fairness in recommendation systems has garnered widespread attention. Moreover, numerous fair recommendation models have been developed for scenarios with limited sensitive information, thereby alleviating the issue of missing sensitive information. However, the performance of these methods still tends to decline significantly when sensitive attributes are extremely scarce. In this paper, we propose FairCL, a novel fair recommendation framework designed to perform effectively under limited sensitive attribute information. FairCL features a contrastive learning-based sensitive attribute encoder that can be integrated with existing fair recommendation algorithms. By leveraging both collaborative information and item side information, we predict unknown sensitive attributes and apply contrastive learning for sensitive attribute modeling. Furthermore, we theoretically demonstrate how FairCL can be integrated with mutual information-based and adversarial learning-based fairness algorithms. Extensive experiments on three real-world datasets show that FairCL significantly enhances fairness, even when only a small portion of users' sensitive attributes are known. The code and data are at: https://anonymous.4open.science/r/CL-for-FairRec-000A/.

With the widespread adoption of mobile devices, an increasing number of users are engaging in social interactions through these devices. Mobile social networks have thus emerged in response. Existing research has shown that utilizing mobile social relations can effectively enhance the performance of recommendation systems. However, most studies only exploit single social relations such as pairwise relations, overlooking the effect of high-order complexity of user relations which contain some potentially beneficial information. What's more, they ignore the impact of the trusters who provide some potential feedbacks in the mobile social network. Therefore, this paper proposes our framework H2TRec using hypergraph convolution in the pretraining stage to learn high-order neighbor information in the mobile social network and utilizing the metapath-based GAT to model users' bidirectional trust relations. First, mobile social communities are partitioned by random walk based on the fusion graph which consolidates all different nodes and relations. Second, each mobile social community is represented as a hyperedge to construct the hypergraph and the high-order neighbor prior knowledge is learned using hypergraph convolution. Third, implicit relations are mined and the metapath-based GAT is utilized to model the preferences of users and items. Notably, unlike previous work, we consider mobile users' out-degree and in-degree features, which enhance the user embeddings. Additionally, a loss term aiming to improve centrality is added to make the preference features of mobile social communities more prominent. Extensive experiments on five popular real-world datasets demonstrate that our H2TRec can improve precision compared with state-of-the-art methods. We release the source code athttps://github.com/kangkang-yun/H2TRec.

Accurate long-term traffic forecasting improves traffic efficiency and safety in Intelligent Transportation Systems (ITS). However, previous methods typically overlook the complex mixed characteristics. So they struggle to capture the intricate features of newly added data effectively under distribution shifts, which exacerbates the complexity of spatial-temporal variations. Moreover, these methods fail to model global and local dynamic spatial correlations effectively, efficiently, and comprehensively. To mitigate these issues, this paper proposes an innovative Spatial-Temporal Wavelet Attention Network (STWAN). STWAN first decomposes traffic data into stable trends and fluctuating events, effectively dealing with the adverse effects of distribution shifts. Then, spatial-temporal encoder captures component-specific temporal variations and extracts dynamic global-local spatial correlations comprehensively with linear computational complexity. Additionally, transformer attention and forecasting decoder model the latent patterns, while trend-event fusion module integrates essential information for accurate forecasts. Comprehensive experiments across two real-world traffic forecasting tasks indicate that STWAN significantly outperforms state-of-the-art methods in terms of accuracy and robustness, attaining a maximum reduction of 5.54% in MAE and showcasing its robustness and broad applicability in long-term forecasting.

Wireless Sensor Network (WSN) reliability evaluation is essential for ensuring the stable operation of network. Traditional methods usually focus on the network topology structure, and calculate the normal operation probability of WSNs. However, these methods usually ignore the energy consumption and network lifetime. In this paper, a novel reliability evaluation algorithm TLR is proposed, which calculates the network lifetime under dynamic network environment according to the pre-set network topology structure reliability threshold, and realizes the comprehensive reliability analysis of network topology and lifetime. In addition, as the basis for reliability evaluation, this paper proposes a new deep reinforcement learning network framework GNN-AC combining graph neural network and actor-critic network, which solves the challenge of constructing Virtual Backbone Network (VBN) in dynamically operating networks. Based on the self-defined fitness matrix and fitness value, the objective function is set to optimize the VBN construction scheme to accurately calculate the network lifetime, and the relationship between the reliability of network topology and network lifetime is discussed. Simulations are carried out for various sizes of WSNs to show the advantages and effectiveness of the proposed approach in estimating network lifetime and reliability evaluation.

Artificial Intelligence-empowered Industrial Control Networks coordinate massive heterogeneous devices and contain multi-node spatial-temporal information. Multivariate Time Series Anomaly Detection (MTS-AD) can discover data-fault behaviors for ensuring the security of distributed networks. However, existing studies tend to rely heavily on single temporal features or neglect the rich spatial-temporal correlations, which leads to the serious underutilization of interactive embeddings between the time and space domains. In this article, a novel Multiview Spatial-Temporal Interaction Attention Network (MSTIA-Net) scheme is proposed for the unsupervised MTS-AD task to better tackle these challenges. MSTIA-Net focuses on jointly modeling the comprehensive spatial-temporal dependencies by means of incorporating complex interactive contents and dynamic relations from multiview patterns. To fully leverage the content-oriented interactions, a spatial-temporal interactions aggregation module is presented to explicitly learn content-aware representations with a parallel-attention mechanism and a low-rank bilinear fusion manner. Simultaneously, considering the potential correlations among different variables as contextual cues, a spatial-temporal correlations learning module is developed to adaptively capture the relevant context for relation-aware representations. On this basis, both types of aware clues are further integrated by the dual attention-enhanced contrastive reconstruction, which can enrich the cross-aware fusion representations and generate the local and global outputs through a cross-view contrastive learning strategy. Experiments conducted on six benchmark datasets demonstrate the superiority of our MSTIA-Net over state-of-the-art baselines.

Network quantization, one of the most widely studied model compression methods, effectively quantizes a floating-point model to obtain a fixed-point one with negligible accuracy loss. Although great success was achieved in reducing the model size, it may exacerbate the unfairness in model accuracy across different groups of datasets. This paper considers two widely used algorithms: Post-Training Quantization (PTQ) and Quantization-Aware Training (QAT), with an attempt to understand how they cause this critical issue. Theoretical analysis with empirical verifications reveals two responsible factors, as well as how they influence a metric of fairness in depth. A comparison between PTQ and QAT is then made, explaining an observation that QAT behaves even worse than PTQ in fairness, although it often preserves a higher accuracy at lower bit-widths in quantization. Finally, the paper finds out that several simple data augmentation methods can be adopted to alleviate the disparate impacts of quantization, based on a further observation that class imbalance produces distinct values of the aforementioned factors among different attribute classes. We experiment on either imbalanced (UTK-Face and FER2013) or balanced (CIFAR-10 and MNIST) datasets using ResNet and VGG models for empirical evaluation.

We explore the potential of \emph{AI-enhanced combinatorial optimization theory}, taking online bipartite matching (OBM) as a case study. In the theoretical study of OBM, the \emph{hardness} corresponds to a performance \emph{upper bound} of a specific online algorithm or any possible online algorithms. Typically, these upper bounds derive from challenging instances meticulously designed by theoretical computer scientists. Zhang et al. (ICML 2024) recently provide an example demonstrating how reinforcement learning techniques enhance the hardness result of a specific OBM model. Their attempt is inspiring but preliminary. It is unclear whether their methods can be applied to other OBM problems with similar breakthroughs. This paper takes a further step by introducing DiMa, a unified and novel framework that aims at understanding the hardness of OBM problems based on denoising diffusion probabilistic models (DDPMs). DiMa models the process of generating hard instances as denoising steps, and optimizes them by a novel reinforcement learning algorithm, named \emph{shortcut policy gradient} (SPG). We first examine DiMa on the classic OBM problem by reproducing its known hardest input instance in literature. Further, we apply DiMa to two well-known variants of OBM, for which the exact hardness remains an open problem, and we successfully improve their theoretical state-of-the-art upper bounds.

Due to the rapid increase of heterogeneous IoT devices in smart home, it is a challenge to model the communication behaviors of IoT devices for identify and protect vulnerable devices. Recently, much researches focus on identifying IoT devices with traffic feature extraction and machine learning algorithms. However, existing methods typically design complex features involving packet payloads for high identification accuracy, which leaks user privacy and leads to large computational overhead. In this paper, we propose SecFinder, a privacy protection-based system for fine-grained IoT device identification in smart home. First, SecFinder analyzes the characteristics of network traffic from IoT devices and uses the statistical information of packet size from the traffic flow to characterize various devices in order to protect privacy. To reduce computational overhead and improve recognition accuracy, SecFinder divides traffic with fixed-size time window and extract only six statistical features from each flow. Then a multi-stage classifier is applied to identify IoT and non-IoT devices, IoT device types and the detailed information of each IoT device. SecFinder is evaluated with traffic from a public real-world device dataset and five classifiers. The results validate that SecFinder can achieve higher accuracy and lower latency in identification than existing methods.

Coverage maximization under full connection constraints involves many factors and poses huge challenge in mobile wireless sensor networks. Most of research works on this issue are based on the disk model, which have high complexity and long iterations, and therefore cannot be applied to dynamic complex networks. In this paper, the problem of confident information coverage maximization under full connectivity constraints (CIC-CC) is defined based on the confident information coverage model (CIC). To address this problem, a 3-stage connectivity constrained coverage maximization algorithm (3-CCC) is proposed with the time complexity of O(T*n2). 3-CCC contains three stages: maximizing coverage (MC), full connectivity (FC), and maximizing connectivity constrained coverage (MCCC). These three stages can be used in whole or in part to achieve coverage maximization with connectivity constraints depending on the network state. The stable matching mechanism, greedy algorithm, and Q-learning are utilized to improve the algorithm’s efficiency. Experiments show that the proposed algorithm has good performance in terms of iteration number, running time, and coverage rate.

Industrial Internet of Things (IIoT) is evolving toward decentralization and autonomous operation. Nodes in decentralized IIoT collaboratively sense data and communicate to participate in meta-computing and provide diverse intelligent services. Area coverage reliability evaluates the information collaborative intelligence sensing and communication capabilities of nodes in decentralized IIoT for a target area. It serves as a crucial performance evaluation metric for the meta-computing and intelligent services of decentralized IIoT. Existing area coverage reliability evaluation algorithms do not consider the dynamic node states in decentralized IIoT, and overlook the capability of nodes to collaborate in meta-computing. To address these issues, this article proposes a novel confident information coverage (CIC)-based area coverage reliability metric (CACREL), which comprehensively considers node multistate, node collaboration, network coverage rate, and network connectivity. To effectively evaluate CACREL, a CIC-based coverage reliability evaluation algorithm (CACR) is proposed. Specifically, CACR transforms complex networks into grid networks and utilizes a coverage table (CT) to describe different coverage states of each grid, which reduces the complexity of the area coverage reliability evaluation. Additionally, CACR employs a reliable path algorithm to converge the CT of each grid to the sink node based on grid connectivity. Simulation results demonstrate that the proposed CACR can accurately evaluate the area coverage reliability of decentralized IIoT and enhance its service performance.

Wireless sensor network plays a crucial role in sensing everything in Internet of Everything (IoE) applications. Network reliability, which measures the ability of the network to satisfy specific requirements, is one of the core factors influencing the quality of service of the network and a vital support for ensuring the normal operation of IoE applications. Existing reliability evaluation methods are mainly based on minimum cutsets or paths, which are inefficient and not suitable for large-scale networks. Furthermore, most work either focuses on coverage functionality or connectivity functionality, lacking energy awareness. To address these limitations, this article proposes a multidimensional target full-coverage reliability (TFCR). TFCR comprehensively considers various factors affecting network reliability. To evaluate TFCR, a graph-empowered confident information coverage (CIC) and signal-to-interference and noise ratio (SINR)-based energy-aware reliability algorithm (CSERA) is proposed. This algorithm evaluates network coverage based on the CIC model. Additionally, graph neural networks and the SINR-based fade tail connectivity (FTC) model are used to evaluate network connectivity functionality. CSERA balances computational accuracy and efficiency, providing reliability evaluation values within an acceptable margin of error. Extensive simulations and comparative experiments from multiple perspectives demonstrate the superiority of the proposed method CSERA over existing approaches.

The Cloud-Fog Automation is a newly proposed digital industrial automation architecture aimed at accelerating the integration and collaboration of communication, computing, and control towards next-generation cyber-physical systems (CPSs). Data reliability is one of the key considerations for achieving Cloud-Fog Automation. Sensor nodes serve as infrastructures for data collection within industrial CPSs and are essential for maintaining ultra-high data reliability. However, the underlying sensor nodes communicate frequently, are damage-prone and difficult to identify, which dramatically shortens the network lifetime and poses great challenges to data reliability. Motivated by this fact, this paper co-designs communication architecture, algorithms, and computing models for next-generation industrial CPSs with Cloud-Fog Automation to ensure data reliability and functional security. First, a four-layer energy-efficient communication architecture is proposed and a cluster head computing algorithm based on double deep Q-learning (CH-DDQ) is designed inside the architecture. Besides, a 2-stage hyBrid fault detection scheme (2-Brain) is proposed for underlying sensor nodes. 2-Brain first incorporates the Obstacle Triple Jump Protocol (OTP) and OTP packets to improve hard fault detection performance. Then, an unsupervised sensor reading soft fault detection model (SR-SFD) based on contrastive learning, momentum, and tensor is adopted to learn discriminative representations of sensor readings and identify soft faults. Simulations and a case study in the nuclear power industry manifest CH-DDQ improves the network lifetime by 5.4%~484.3% compared to three peer methods, and OTP performs better than baselines by 33.1% on average. Additionally, SR-SFD exhibits high efficiency in sensor soft fault detection and other application scenarios.

As a key infrastructure for social fintech ecosystems, ethereum enables decentralized finance (DeFi) applications where security issues directly compromise ecosystem stability. Among critical security concerns, ethereum phishing scams stand as typical scams. Criminals employ distinctive fund transfer patterns (e.g., money laundering stages: placement, layering, and integration) to obscure illicit funds through long transaction paths. While graph neural networks (GNNs) dominate detection methods, they fail to model these long paths effectively. To address this, we propose the first framework to detect phishing scams through explicitly modeling fund transfer patterns. Our novel method, IMPUTATION, introduces: 1) a heuristic fund transfer path graph construction method utilizing iterative transaction pairing to capture complicated fund transfer patterns; 2) role-topology account embeddings encoding fund transfer patterns; 3) attention fusion leveraging initial transactions to suppress path noise; and 4) heterogeneous correlation graphs with weighted adjacency reconstruction modeling interpath dependencies. Extensive experiments demonstrate that IMPUTATION outperforms on all five metrics and detecting Ethereum phishing scams from fund transfer patterns is effective.

Empowered by the continuous integration of social multimedia and artificial intelligence, the application scenarios of Information Retrieval (IR) progressively tend to be diversified and personalized. Currently, User-Generated Content (UGC) systems have great potential to handle the interactions between large-scale users and massive media contents. As an emerging multimedia IR, Fashion Compatibility Modeling (FCM) aims to predict the matching degree of each given outfit and provide complementary item recommendation for user queries. Although existing studies attempt to explore the FCM task from a multi-modal perspective with promising progress, they still fail to fully leverage the interactions between multi-modal information or ignore the item–item contextual connectivities of intra-outfit. In this article, a novel FCM scheme is proposed based on Correlation-Aware Cross-Modal Attention Network. To better tackle these issues, our work mainly focuses on enhancing comprehensive multi-modal representations of fashion items by integrating the cross-modal collaborative contents and uncovering the contextual correlations. Since the multi-modal information of fashion items can deliver various semantic clues from multiple aspects, a modality-driven collaborative learning module is presented to explicitly model the interactions of modal consistency and complementarity via a co-attention mechanism. Considering the rich connections among numerous items in each outfit as contextual cues, a correlation-aware information aggregation module is further designed to adaptively capture significant intra-correlations of item–item for characterizing the content-aware outfit representations. Experiments conducted on two real-world fashion datasets demonstrate the superiority of our approach over state-of-the-art methods.

AI-empowered edge computing has given rise to a new paradigm and effectively facilitated the promotion and development of multimedia applications. The speech assistant is one of the significant services provided by multimedia applications, which aims to offer intelligent interactive experiences between humans and machines. However, malicious attackers may exploit spoofed speeches to deceive speech assistants, posing great challenges to the security of multimedia applications. The limited resources of multimedia terminal devices hinder their ability to effectively load speech spoofing detection models. Furthermore, processing and analyzing speech in the cloud can result in poor real-time performance and potential privacy risks. Existing speech spoofing detection methods rely heavily on annotated data and exhibit poor generalization capabilities for unseen spoofed speeches. To address these challenges, this article first proposes the Coordinate Attention Network (CA2Net) that consists of coordinate attention blocks and Res2Net blocks. CA2Net can simultaneously extract temporal and spectral speech feature information and represent multi-scale speech features at a granularity level. Besides, a contrastive learning-based speech spoofing detection framework named GEMINI is proposed. GEMINI can be effectively deployed on edge nodes and autonomously learn speech features with strong generalization capabilities. GEMINI first performs data augmentation on speech signals and extracts conventional acoustic features to enhance the feature robustness. Subsequently, GEMINI utilizes the proposed CA2Net to further explore the discriminative speech features. Then, a tensor-based multi-attention comparison model is employed to maximize the consistency between speech contexts. GEMINI continuously updates CA2Net with contrastive learning, which enables CA2Net to effectively represent speech signals and accurately detect spoofed speeches. Extensive experiments on the ASVspoof2019 dataset show that GEMINI reduces the Equal Error Rate and tandem Detection Cost Function by up to 96.75% and 96.35% in the physical access scenario, and by up to 86.62% and 87.71% in the logical access scenario compared to peer methods.

Voice assistants, as crucial components of the Internet of Things (IoT), are vulnerable to voice spoofing attacks and pose great threats to the security of IoT. Passive liveness detection distinguishes between genuine and spoofing voices by analyzing the collected voice, eliminating the need for deploying additional sensors. This method plays a crucial role in detecting spoofing speeches and ensuring the security of the IoT. However, current passive liveness detection methods typically require users to adopt specific gestures. Meanwhile, these methods are often designed for specific attacks and cannot accommodate multivariate attacks. To address these challenges, this paper proposes an efficient and robust liveness feature called VoiceID, which utilizes the inherent vocal cord vibrations and voiced language to authenticate the collected voice. The VoiceID is defined as the set of maximum magnitude-peak frequency bins in the magnitude spectrum of each frame for voice. VoiceID can be combined with existing acoustic features to compensate for the granularity gap in extracting fine-grained features and distinguishing between genuine and spoofing voices. Furthermore, to leverage VoiceID, this paper proposes a solid fake voice liveness detection system named SFSys and elaborates on a series of acoustic features that can work with VoiceID. Extensive experiments on authoritative ASVspoof 2019 and ASVspoof 2021 datasets reveal that VoiceID reduces the equal error rate and the minimum tandem decision cost function of the existing acoustic features by at most 6.19% and 0.2479. Moreover, SFSys outperforms existing voice liveness detection schemes and exhibits robustness in various advanced spoofing attack environments.

Introducing a Knowledge Graph (KG) to facilitate a recommender system has become a tendency in recent years. Many existing methods leverage KGs to obtain side information of items to promote item representation learning for enhancing recommendation performance. However, they ignore that KGs also may contribute to better user representation learning. To solve this issue, we propose a novel algorithm, the KIGR ( K nowledge-aware I nteraction G raph for R ecommendation), to mine user–item interactions via KGs for assisting user representation learning. Specifically, a user–item interaction is encoded by attentively summing up the relation embedding about the item in the KG. Then, an unsupervised learning method is used to group the user–item interactions into different latent types. Further, a user–item interaction graph is divided into several subgraphs, which is referred to as a Knowledge-aware Interaction Graph, making each subgraph only contain one latent type of interaction. Finally, user representation is the fusion of user interest embedding, which is learned on the knowledge-aware interaction graph, whereas item representation is learned on the KG. Experimental results on MovieLens, LastFM and Amazon-Book validate that the proposed KIGR has a superior performance compared with the state-of-the-art algorithms.

Internet of Things (IoT) reliability evaluation contributes to the sustainable computing and enhanced stability of the network. Previous algorithms usually evaluate the reliability of IoT by enumenating the states of nodes and networks, which are difficult to handle IoT with hundreds of nodes because the computational cost. In this paper, a novel algorithm, TMCRA, is proposed to evaluate the reliability of IoT in complex network environment, which consider both coverage and connectivity. For coverage, TMCRA employs the Confident Information Coverage (CIC) model to divide the target area into independent grids and calculates the coverage rate. In terms of connectivity, TMCRA forming the Virtual Backbone Network (VBN) based on two proposed methods: TMA and MGIN, and evaluate connectivity by analyzing the VBN rather than the whole network. The TMA and MGIN are two algorithms for constructing Minimum Connected Dominant Sets (MCDS), which are suitable for different scale networks. Finally, based on the data of coverage and connectivity, TMCRA utilizes tensors for the unified modeling and representation of network structure, and calculates IoT reliability based on the tensors. Simulations are carried out for various sizes of IoT to show the advantages and effectiveness of the proposed approach in reliability evaluation.

Multimodal image pairs can make object detection more reliable in challenging environments, so RGB-T object detection has gained extensive attention over the past decade. To alleviate the complementarity of the visible and thermal modality, we propose a novel lightweight Feature Enhancement-fusion Module (FEM), which is composed of the Channel Enhancement-fusion Unit (CEU) and Spatial Enhancement-fusion Unit (SEU) by extending the attention mechanism to operate on two modalities. CEU is used to exploit the complementarity and alleviate the data imbalance by combining internal and global channel attention. Additionally, SEU is utilized to guide the model to pay more attention to the regions of interest. By incorporating FEM, enhanced and fused features are obtained, leading to improved performance. The effectiveness and generalizability of FEM are validated by two public datasets and our proposed DCSANet achieves competitive performance while maintaining high speed (+%7.0 on LLVIP and +1.2% on FLIR in mAP). Moreover, we conducted ablation experiments to verify the effectiveness of the proposed operators.

In order to resist network attacks on IoT devices, identifying IoT devices is the first step for ensuring device security. The traditional passive method identifies IoT devices by mining the potential relationship between traffic characteristics and devices. However, the form of selected traffic features are too singular without considering device behavioral characteristics and the classifier used is too specific with simple structure in these methods. This paper proposes a stacking ensemble learning approach for IoT device identification, ENSIOT, which fully considering the behavioral characteristics of devices and integrating the advantages of various machine learning methods to achieve efficient identification of IoT devices. Firstly, in the process of traffic processing, our method selects features from activity cycles, port numbers, signalling patterns, and cipher suites. Then, in model integration, many machine learning methods are used as base models to learn features selected, and output preliminary recognition results. Finally, the meta model learns the relationship between label and the recognition results of each base model and outputs the final device identification result. This stacking structure stacks the base models and the meta model to make a classifier with strong identification and generalization ability. Incremental learning is used to improve identification accuracy when traffic pattern changing. Comparative experiments are conducted on two datasets of UNSW and TMA-2021. The experimental results verify the effectiveness of ENSIOT, which achieve the accuracy of over 98% on two dataset and bring a noticeable improvement in terms of both accuracy and macro F1 score.

\beginabstract Social network has obtained extensive attention in recommender system. Existing social recommendation models mostly leverage social relations to capture potential interactions between users and items, thereby enhancing recommendation performance. However, these methods ignore the fine-grained bidirectional trust weight and the constraint on the relative positions of entities in social network and user-item interaction network. To this end, in this paper, we propose a social recommendation framework with Trust mining and multi-semantic reGularization (TrustGo). Specifically, we firstly construct a trust network based on the observed social network and establish a high-quality item implicit network. Then, we integrate the trust network, item implicit network, and user-item interaction network into a heterogeneous network. We introduce a meta-path based aggregation in this heterogeneous network to map the users and items into a latent space. And then, by using an ensemble method, we can obtain the final prediction ratings. Considering the users' different behaviors in social network and user-item interaction network, we define two semantic spaces, i.e., the social semantic space and user-item interactional semantic space. And a multi-semantic regularization module is designed to adjust the relative positions of entities in the two kinds of semantic spaces, respectively. Extensive experiments on three real-world datasets demonstrate that our TrustGo model is superior to other state-of-the-art recommendation models. \endabstract

The continuous combination of digital network technology and traditional financial services has given birth to digital financial networks, which explore massive economic data under the AI-driven models to achieve intelligent connections among financial institutions, markets, transactions, and instruments. Empirical asset pricing is a challenging task in financial analysis, which has attracted research attention. However, existing studies only focus on tackling the challenges of equity risk premium in the single stock market. Considering multiple economic linkages between the two countries, the transaction history of the US stock market as empirical knowledge is a powerful supplement to improve the prediction of equity risk premium in the China market. In this paper, we aim to fully leverage the prior information in two stock markets for empirical asset pricing models. Due to the rich financial domain knowledge, there may be various characteristic signals that partially overlap in different periods. To address these issues, we propose a framework based on long-short dual-mode knowledge distillation, termed as LSDM-KD, which incorporates US and China stock market models, and a shared characteristic signals model. The method effectively understands the relationships between assets and market behaviour, reducing reliance on expensive correlation databases and professional knowledge. Extensive experiments conducted on US and China stock market datasets demonstrate that our LSDM-KD can significantly improve the performance of empirical asset pricing.

With the development of recommendation algorithms, researchers are paying increasing attention to fairness issues such as user discrimination in recommendations. To address these issues, existing works often filter users’ sensitive information that may cause discrimination during the process of learning user representations. However, these approaches overlook the latent relationship between items’ content attributes and users’ sensitive information. In this article, we propose DALFRec, a fairness-aware recommendation algorithm based on user-side and item-side adversarial learning to mitigate the effects of sensitive information on both sides of the recommendation process. First, we conduct a statistical analysis to demonstrate the latent relationship between items’ information and users’ sensitive attributes. Then, we design a dual-side adversarial learning network that simultaneously filters out users’ sensitive information on the user and item side. Additionally, we propose a new evaluation strategy that leverages the latent relationship between items’ content attributes and users’ sensitive attributes to better assess the algorithm’s ability to reduce discrimination. Our experiments on three real datasets demonstrate the superiority of our proposed algorithm over state-of-the-art methods.

Discrimination against different user groups has received growing attention in the recommendation field. To address this problem, existing works typically remove sensitive attributes that may cause discrimination through adversary learning to achieve fair recommendations. However, these approaches leverage all available interactions for learning user representations and overlook the fact that different interactions have varying relevance to users’ sensitive attributes. Ignoring this issue may weaken the effectiveness of adversary learning in removing sensitive attributes. To tackle this challenge, we propose a novel model called GS-FairRec, which distinguishes between user interactions to achieve better removal of sensitive attributes. The model consists of three modules: graph sampling-based representation learning, pseudo-user representation learning, and adversarial learning. Firstly, the graph sampling-based representation learning module removes some irrelevant neighbors from a user-item bipartite graph and employs a graph convolutional network (GCN) to learn user/item representations. Next, items that are relevant to a user’s sensitive information but do not match their preferences are defined as the user’s pseudo-interest items, which are leveraged to learn the pseudo-user representation. In the adversarial learning module, the user’s two kinds of representations are fused for adversarial learning to remove sensitive information. Additionally, we design a new metric to measure the model’s ability to remove sensitive attributes based on how a generated recommendation list discloses the user’s sensitive attributes. Finally, we conduct experiments on two real-world datasets, and our results demonstrate the superiority of our proposed model in fairness tasks.

Incorporating knowledge graph as side information has become a new trend in recommendation systems. Recent studies regard items as entities of a knowledge graph and leverage graph neural networks to assist item encoding, yet by considering each relation type independently. However, relation types are often too many and sometimes one relation type involves too few entities. We argue that there may exist some latent relevance among relations in KG. It may not necessary nor effective to consider all relation types for item encoding. In this paper, we propose a VRKG4Rec model (Virtual Relational Knowledge Graphs for Recommendation), which clusters relations with latent relevance to generates virtual relations. Specifically, we first construct virtual relational graphs (VRKGs) by an unsupervised learning scheme. We also design a local weighted smoothing (LWS) mechanism for node encoding on VRKGs, which iteratively updates a node embedding only depending on the node itself and its neighbors, but involve no additional training parameters. LWS mechanism is also employed on a user-item bipartite graph for user representation learning, which utilizes item encodings with virtual relational knowledge to help train user representations. Experiment results on two public datasets validate that our VRKG4Rec model outperforms the state-of-the-art methods. The implementations are available at https://github.com/lulu0913/VRKG4Rec.

The fairness consideration has received increasing attention in artificial intelligence (AI), especially in the significant application, recommender system. One kind of fairness issue is to balance the exposure of popular items and less popular items. Existing works mainly compensate the less popular items when predicting their ratings. However, the compensation for the items may result in a loss of recommendation accuracy, as the compensated items may not match user’s preference. For this problem, we propose a multi-community clustering recommendation with fair decision fusion (MultiFDF) framework to compensate less popular items locally, which can reduce the negative impact on user’s preferred popular item. We first design an experiment and deliver a causal graph-based mathematical proof to demonstrate the feasibility of local compensation. It proves that there is a discrete tendency of ratings predicted by a recommendation algorithm and a recommendation algorithm gives a higher rating to the popular item than the less popular item. The MultiFDF consists of three parts, community exploration module, local recommendation module, and fair decision fusion module. The community exploration module outputs several communities for local recommendation module to generate local recommendation lists, respectively. The fair decision fusion module then computes the discrete ratings of items based on local recommendation lists and designs an edge reranking strategy based on their discrete ratings to obtain the final fair top-$N$recommendation list. To verify the superiority of our proposed MultiFDF, we conduct experiments on three real world datasets and the results demonstrate that MultiFDF can improve fairness at the cost of lower accuracy than the state-of the-art algorithms.

Writer retrieval is crucial in document forensics and historical document analysis. However, due to the difference in syntactic structure between Chinese and other languages, the existing methods may not be directly applied to Chinese writer retrieval. Previous work on Chinese writer retrieval does not overcome the performance degradation problem when the number of samples grows. In this paper, we propose a novel compound uniform encoding algorithm (CUE) for Chinese writer retrieval, which mainly consists of a combined feature extraction module (CFE) and a prototype substitution module (PS). The CFE module combines two complementary features from image filter response and character contour. It counts local symmetries and edge co-occurrence pairs. PS module substitutes the outliers with the class prototypes to alleviate the influence of the outliers. Finally, the weighted Chi-square distance is applied to measure the similarity between writer and text. To verify the superiority of our proposed method, experiments are conducted on four public datasets and our built dataset. The results validate that CUE outperforms the state-of-the-art algorithms on mAP metric.

The emergence of new applications of Internet of Things (IoT) makes its security and reliability become one of the most concerning issues and requires more breakthroughs. To ensure reliable operation of IoT, network reliability measures are essential for quantifying the performance of such networks. In this paper, we focus on the problem of coverage reliability of IoT intrusion detection systems based on Attack-Defense Game Design. A comprehensive coverage reliability algorithm is proposed based on Monte Carlo simulations. The algorithm employs Byzantine attack and defense ideas to determine network node attributes and uses confident information model to calculate the network coverage area. Furthermore, we propose a system reliability metric based on the analytic hierarchy process method, which takes advantage of node attributes, network coverage and connectivity. The metric is used to compare algorithms in simulated experiments, and a series of simulation comparisons illustrate the superiority and usability of the proposed approach.

Event recommendation has become an important issue in event-based social networks (EBSN). In this paper, we study how to exploit diverse relations in an EBSN as well as individual history preferences to recommend preferred events. We first construct a hybrid graph consisting of different types of nodes to represent available entities in an EBSN. The graph uses explicit relations as edges to connect nodes of different types; while transferring implicit relations of event attributes to interconnect the event nodes. After executing the graph random walking, we obtain the candidate events with high convergency probabilities. We next extract a user preference from his attended events to further compute his interest similarities to his candidate events. The recommended event list is then obtained by combining the two similarity scores. Data sets from a real EBSN are used to examine the proposed scheme, and experiment results validate its superiority over peer schemes.

This paper describes a fully differential inductor-less temperature independent laser driver for a Mach-Zehnder modulator (MZM). With push pull structure in output stage and adaptive feedback topology in pre-driver stages, the laser driver purposed (fabricated using a 130nm CMOS process) in this work can achieve differential output voltage swing of up to 4Vpp when driving a 50 Ohm load at speeds of up to 12.5Gb/s across a temperature range 27°C to 135°C.