Search

The financial transaction relationships between existing entities are complex and diverse. In this situation, traditional risk control methods mainly ignored such complex and implicit relationship characteristics, remaining difficult to cope with complex and ever-changing financial risks. To address this issue, this paper proposes a novel relational graph convolution network (GCN)-based smart risk recognition model for financial transactions. Firstly, the classic GCN is simplified based on spatiotemporal effect. Then, feature extraction is conducted for financial transaction data, and a transformer encoder-based GCN model is proposed for risk recognition. The proposed model in this work is named as graph transformer graph convolutional network (GT-GCN) for short. In addition, fuzzy evaluation method is added into it. Finally, some experiments are conducted on real-world financial transaction data to make validation for the proposed GT-GCN. The research results indicate that the GT-GCN can not only effectively identify risks in financial transactions, but also has high accuracy and predictive ability. The application of GT-GCN to actual datasets also has good scalability and adaptability, and it can be resiliently extended into many other fields. [ABSTRACT FROM AUTHOR]

As a powerful tool for learning high-dimensional data representation, graph neural networks (GNN) have been applied to predict the remaining useful life (RUL) of rolling bearings. Existing GNN-based RUL prediction methods predominantly rely on constant pre-constructed graphs. However, the degradation of bearings is a dynamic process, and the dependence information between features may change at different moments of degradation. This article introduces a method for RUL prediction based on dynamic graph spatial-temporal dependence information extraction. The raw signal is segmented into multiple periods, and multiple features of each period data are extracted. Then, the correlation coefficient analysis is conducted, and the feature connection graph of each period is constructed based on different analytical results, thereby dynamically mapping the degradation process. The graph data is fed into graph convolutional networks (GCN) to extract spatial dependence between the graph node features in different periods. To make up for the shortcomings of GCN in temporal dependence extraction, the TimesNet module is introduced. TimesNet considers the two-dimensional changes of time series data and can extract the temporal dependence of graph data within and between different time cycles. Experimental results based on the PHM2012 dataset show that the average RUL prediction error of the proposed method is 17.4%, outperforming other comparative methods. [ABSTRACT FROM AUTHOR]

Skeleton-based human action recognition has great potential for human behavior analysis owing to its simplicity and robustness in varying environments. This paper presents a spatial and temporal attention-enhanced graph convolution network (STAEGCN) for human action recognition. The spatial-temporal attention module in the network uses convolution embedding for positional information and adopts multi-head self-attention mechanism to extract spatial and temporal attention separately from the input series of the skeleton. The spatial and temporal attention are then concatenated into an entire attention map according to a specific ratio. The proposed spatial and temporal attention module was integrated with an adaptive graph convolution network to form the backbone of STAEGCN. Based on STAEGCN, a two-stream skeleton-based human action recognition model was trained and evaluated. The model performed better on both NTU RGB+D and Kinetics 400 than 2s-AGCN and its variants. It was proven that the strategy of decoupling spatial and temporal attention and combining them in a flexible way helps improve the performance of graph convolution networks in skeleton-based human action recognition. [ABSTRACT FROM AUTHOR]

Group recommendation involves suggesting items or activities to a group of users based on their collective preferences or characteristics. Graph contrastive learning is a technique used to learn representations of items and users in a graph structure. Although contrastive learning-based recommendation techniques reduce the data sparsity problem by extracting general features from raw data and also make the representation of user-item bipartite graph augmentations more consistent, the factors contributing to improving the performance of this technique are still not fully understood. Meanwhile, graph augmentations have little importance in contrastive learning-based recommendation and are relatively unreliable. The eXtremely Simple Graph Contrastive Learning (XSimGCL) provides novel insights into the effect of contrastive learning on recommendation, where views for contrastive learning are created through a simple yet effective noise-based embedding augmentation. Although XSimGCL infers the final group decision by dynamically aggregating the preferences of group members and includes various types of interaction, the performance of supervised learning is reduced due to the data sparsity problem, and as a result, the efficiency of group preference representation is limited. To address this challenge, we developed a Group Recommendation model based on XsimGCL in this study (GR-GCL). GR-GCL is inspired by the Light Graph Convolution Network (LightGCN) to realize simultaneous learning of multiple graphs, where initial embedding is considered the only update parameter. Also, GR-GCL improves group recommendation by applying cross-layer contrastive learning in the XSimGCL model by representing more diverse entities. The rationality analysis of our proposed GR-GCL has been performed on several datasets from both analytical and empirical perspectives. Although our model is very simple, it performs better in group recommendations by adjusting the uniformity of representations learned from counterparts based on contrastive learning. [ABSTRACT FROM AUTHOR]

The key of current research on behavior recognition algorithms based on deep learning lies in enhancing the accuracy and stability of key point extraction, in order to achieve more accurate action recognition of targets. However, many current algorithms tend to just add attention mechanisms that appear to perform better in the feature extraction stage of the target, without considering the impact of different attention mechanisms on different models and tasks. Therefore, this study proposes an algorithmic model for pose estimation based on various attention mechanisms, which further highlights the importance of selecting an appropriate attention mechanism by comparing the impact of different attention mechanisms on the model. In addition, considering the stability of key point extraction, the initialization of the model is fine tuned to select a more suitable initialization method that improves the performance by increasing the category of weights on network layer judgments. Compared with the performance of the benchmark network model, the model enhances all evaluation metrics on both multiscale and no multiscale CrowdPose datasets, where the average accuracy improvement in both cases is more than 1%. [ABSTRACT FROM AUTHOR]

In the field of autonomous driving, trajectory prediction of traffic agents is an important and challenging problem. Fully capturing the complex spatio-temporal features in trajectory data is crucial for accurate trajectory prediction. This paper proposes a trajectory prediction model called multi-dimensional spatio-temporal feature fusion (MDSTF), which integrates multi-dimensional spatio-temporal features to model the trajectory information of traffic agents. In the spatial dimension, we employ graph convolutional networks (GCN) to capture the local spatial features of traffic agents, spatial attention mechanism to capture the global spatial features, and LSTM combined with spatial attention to capture the full-process spatial features of traffic agents. Subsequently, these three spatial features are fused using a gate fusion mechanism. Moreover, during the modeling of the full-process spatial features, LSTM is capable of capturing short-term temporal dependencies in the trajectory information of traffic agents. In the temporal dimension, we utilize a Transformer-based encoder to extract long-term temporal dependencies in the trajectory information of traffic agents, which are then fused with the short-term temporal dependencies captured by LSTM. Finally, we employ two temporal convolutional networks (TCN) to predict trajectories based on the fused spatio-temporal features. Experimental results on the ApolloScape trajectory dataset demonstrate that our proposed method outperforms state-of-the-art methods in terms of weighted sum of average displacement error (WSADE) and weighted sum of final displacement error (WSFDE) metrics. Compared to the best baseline model (S2TNet), our method achieves reductions of 4.37% and 6.23% respectively in these metrics. [ABSTRACT FROM AUTHOR]

Cervical cancer is a prevalent disease affecting the cervix cells in women and is one of the leading causes of mortality for women globally. The Pap smear test determines the risk of cervical cancer by detecting abnormal cervix cells. Early detection and diagnosis of this cancer can effectively increase the patient's survival rate. The advent of artificial intelligence facilitates the development of automated computer-assisted cervical cancer diagnostic systems, which are widely used to enhance cancer screening. This study emphasizes the segmentation and classification of various cervical cancer cell types. An intuitive but effective segmentation technique is used to segment the nucleus and cytoplasm from histopathological cell images. Additionally, handcrafted features include different properties of the cells generated from the distinct cervical cytoplasm and nucleus area. Two feature rankings techniques are conducted to evaluate this study's significant feature set. Feature analysis identifies the critical pathological properties of cervical cells and then divides them into 30, 40, and 50 sets of diagnostic features. Furthermore, a graph dataset is constructed using the strongest correlated features, prioritizes the relationship between the features, and a robust graph convolution network (GCN) is introduced to efficiently predict the cervical cell types. The proposed model obtains a sublime accuracy of 99.11% for the 40-feature set of the SipakMed dataset. This study outperforms the existing study, performing both segmentation and classification simultaneously, conducting an in-depth feature analysis, attaining maximum accuracy efficiently, and ensuring the interpretability of the proposed model. To validate the model's outcome, we tested it on the Herlev dataset and highlighted its robustness by attaining an accuracy of 98.18%. The results of this proposed methodology demonstrate the dependability of this study effectively, detecting cervical cancer in its early stages and upholding the significance of the lives of women. [ABSTRACT FROM AUTHOR]

Existing research extensively incorporates syntactic information, especially dependency trees, to enhance the performance of relation extraction tasks. However, relying solely on dependency information may not fully exploit the rich semantic and syntactic information contained in sentences, and not all information in dependency trees is substantively helpful for relation extraction. Therefore, this paper proposes the Dependent-Position Relation Graph Convolutional network with Hierarchical Attention (DPR-GHA) for relation extraction, a method that integrates dependency relations and position relations into a hierarchical attention mechanism to effectively capture the relations between entities in text. The method aims to capture rich semantic information and enhance the performance of relation extraction. Specifically, we introduce the dependency relations of sentences and position relations of words to model global dependencies and local features, respectively. Subsequently, a novel hierarchical attention mechanism is introduced into the Graph Convolutional Network (GCN), dynamically adjusting the weights between nodes based on the input of the graph convolutional layer. This adaptive information aggregation enables each node to aggregate information adaptively according to its context and the importance of neighboring nodes. The research results on the SemEval-2010 Task 8 and KBP37 datasets thoroughly validate the effectiveness of the proposed model, demonstrating its significant performance advantage in relation extraction tasks. [ABSTRACT FROM AUTHOR]

Copy number variation (CNV) refers to the number of copies of a specific sequence in a genome and is a type of chromatin structural variation. The development of the Hi‐C technique has empowered research on the spatial structure of chromatins by capturing interactions between DNA fragments. We utilized machine‐learning methods including the linear transformation model and graph convolutional network (GCN) to detect CNV events from Hi‐C data and reveal how CNV is related to three‐dimensional interactions between genomic fragments in terms of the one‐dimensional read count signal and features of the chromatin structure. The experimental results demonstrated a specific linear relation between the Hi‐C read count and CNV for each chromosome that can be well qualified by the linear transformation model. In addition, the GCN‐based model could accurately extract features of the spatial structure from Hi‐C data and infer the corresponding CNV across different chromosomes in a cancer cell line. We performed a series of experiments including dimension reduction, transfer learning, and Hi‐C data perturbation to comprehensively evaluate the utility and robustness of the GCN‐based model. This work can provide a benchmark for using machine learning to infer CNV from Hi‐C data and serves as a necessary foundation for deeper understanding of the relationship between Hi‐C data and CNV. [ABSTRACT FROM AUTHOR]

Prediction of Remaining Useful Life (RUL) for Rolling Element Bearings (REB) has attracted widespread attention from academia and industry. However, there are still several bottlenecks, including the effective utilization of multi-sensor data, the interpretability of prediction models, and the prediction across the entire life cycle, which limit prediction accuracy. In view of that, we propose a knowledge-based explainable life-cycle RUL prediction framework. First, considering the feature fusion of fast-changing signals, the Pearson correlation coefficient matrix and feature transformation objective function are incorporated to an Improved Graph Convolutional Autoencoder. Furthermore, to integrate the multi-source signals, a Cascaded Multi-head Self-attention Autoencoder with Characteristic Guidance is proposed to construct health indicators. Then, the whole life cycle of REB is divided into different stages based on the Continuous Gradient Recognition with Outlier Detection. With the development of Measurement-based Correction Life Formula and Bidirectional Recursive Gated Dual Attention Unit, accurate life-cycle RUL prediction is achieved. Data from self-designed test rig and PHM 2012 Prognostic challenge datasets are analyzed with the proposed framework and five existing prediction models. Compared with the strongest prediction model among the five, the proposed framework demonstrates significant improvements. For the data from self-designed test rig, there is a 1.66 % enhancement in Corrected Cumulative Relative Accuracy (CCRA) and a 49.00 % improvement in Coefficient of Determination (R2). For the PHM 2012 datasets, there is a 4.04 % increase in CCRA and a 120.72 % boost in R2. [Display omitted] • Advocate explainable knowledge-data integration for RUL prediction throughout the life cycle, leveraging multi-sensor data. • Introduce a novel unsupervised learning framework that integrates IGCA and CMSACG to construct HI. • A two-stage prediction framework, fusing MCLF and BR-GDAU, is proposed for predicting the life cycle RUL. • The effectiveness of the proposed framework is validated using both Self-Designed Experiment-Derived and PHM 2012 Prognostic Challenge Bearing Datasets. [ABSTRACT FROM AUTHOR]

Online social networks (OSNs) are an indispensable part of social communication where people connect and share information. Spammers and other malicious actors use the OSN's power to propagate spam content. In an OSN with mutual relations between nodes, two kinds of spammer detection methods can be employed: feature based and propagation based. However, both of these are incomplete in themselves. The feature-based methods cannot exploit mutual connections between nodes, and propagation-based methods cannot utilize the rich discriminating node features. We propose a hybrid model—Markov enhanced graph attention network (MEGAT)—using graph attention networks (GAT) and pairwise Markov random fields (pMRF) for the spammer detection task. It efficiently utilizes node features as well as propagation information. We experiment our GAT model with a smoother Swish activation function having non-monotonic derivatives, instead of the leakyReLU function. The experiments performed on a real-world Twitter Social Honeypot (TwitterSH) benchmark dataset and subsequent comparative analysis reveal that our proposed MEGAT model outperforms the state-of-the-art models in accuracy, precision–recall area under curve (PRAUC), and F1-score performance measures. [ABSTRACT FROM AUTHOR]

Studying gene regulatory networks (GRNs) is paramount for unraveling the complexities of biological processes and their associated disorders, such as diabetes, cancer, and Alzheimer's disease. Recent advancements in computational biology have aimed to enhance the inference of GRNs from gene expression data, a non-trivial task given the networks' intricate nature. The challenge lies in accurately identifying the myriad interactions among transcription factors and target genes, which govern cellular functions. This research introduces a cutting-edge technique, EGRC (Effective GRN Inference applying Graph Convolution with Self-Attention Graph Pooling), which innovatively conceptualizes GRN reconstruction as a graph classification problem, where the task is to discern the links within subgraphs that encapsulate pairs of nodes. By leveraging Spearman's correlation, we generate potential subgraphs that bring nonlinear associations between transcription factors and their targets to light. We use mutual information to enhance this, capturing a broader spectrum of gene interactions. Our methodology bifurcates these subgraphs into 'Positive' and 'Negative' categories. 'Positive' subgraphs are those where a transcription factor and its target gene are connected, including interactions among their neighbors. 'Negative' subgraphs, conversely, denote pairs without a direct connection. EGRC utilizes dual graph convolution network (GCN) models that exploit node attributes from gene expression profiles and graph embedding techniques to classify these. The performance of EGRC is substantiated by comprehensive evaluations using the DREAM5 datasets. Notably, EGRC attained an AUROC of 0.856 and an AUPR of 0.841 on the E. coli dataset. In contrast, the in silico dataset achieved an AUROC of 0.5058 and an AUPR of 0.958. Furthermore, on the S. cerevisiae dataset, EGRC recorded an AUROC of 0.823 and an AUPR of 0.822. These results underscore the robustness of EGRC in accurately inferring GRNs across various organisms. The advanced performance of EGRC represents a substantial advancement in the field, promising to deepen our comprehension of the intricate biological processes and their implications in both health and disease. [ABSTRACT FROM AUTHOR]

Within the realm of social recommendation, a recommender system can enhance its performance through the use of social information among users. Due to the abundance of redundant information in user interactions and social connections, it affects the performance of recommendation results negatively. Existing recommendation models do not distinguish the influence of different users and different friends. To solve this problem, this paper introduces a new recommendation framework, user preference and social relationship-aware light graph convolutional networks (USLGCN). The proposed framework distinguishes between users based on their interactions with items and social relationships to enhance recommendation accuracy. Specifically, we design a subgraph classification strategy that divides the user–item interaction graph and social graph into different subgraphs to capture the impact of various user types on items and friends, thereby reducing negative information and enhancing model resilience. On top of that, we also design a graph fusion module that enhances recommendation performance by fusing data from multiple subgraphs together. Experiments on public datasets show that USLGCN exhibits a 2.6% increase in recall accuracy compared to other social recommendation methods. [ABSTRACT FROM AUTHOR]

Real-time and accurate traffic flow forecasting plays a crucial role in transportation systems and holds great significance for urban traffic planning, traffic management, traffic control, and more. The most difficult challenge is the extraction of temporal features and spatial correlations of nodes in traffic flow forecasting. Meanwhile, graph convolutional networks has shown good performance in extracting relational spatial dependencies in existing methods. However, it is difficult to accurately mine the hidden spatial-temporal features of the traffic network by using graph convolution alone. In this paper, we propose a multi-attention gated temporal graph convolution network (MATGCN) for accurately forecasting the traffic flow. Firstly, we propose a gated multi-modal temporal convolution(MTCN) to handle the long-term series of the raw traffic data. Then, we use an efficient channel attention module(ECA) to extract temporal features. For the complexity of the spatial structure of traffic roads, we develop multi-attention graph convolution module (MAGCN)including graph convolution and graph attention to further extract the spatial features of a road network. Finally, extensive experiments are carried out on several public traffic datasets, and the experimental results show that our proposed algorithm outperforms the existing methods. [ABSTRACT FROM AUTHOR]

Kitchen waste images encompass a wide range of garbage categories, posing a typical multi-label classification challenge. However, due to the complex background and significant variations in garbage morphology, there is currently limited research on kitchen waste classification. In this paper, we propose a multi-head attention-driven dynamic graph convolution lightweight network for multi-label classification of kitchen waste images. Firstly, we address the issue of large model parameterization in traditional GCN methods by optimizing the backbone network for lightweight model design. Secondly, to overcome performance losses resulting from reduced model parameters, we introduce a multi-head attention mechanism to mitigate feature information loss, enhancing the feature extraction capability of the backbone network in complex scenarios and improving the correlation between graph nodes. Finally, the dynamic graph convolution module is employed to adaptively capture semantic-aware regions, further boosting recognition capabilities. Experiments conducted on our self-constructed multi-label kitchen waste classification dataset MLKW demonstrate that our proposed algorithm achieves a 8.6% and 4.8% improvement in mAP compared to the benchmark GCN-based methods ML-GCN and ADD-GCN, respectively, establishing state-of-the-art performance. Additionally, extensive experiments on two public datasets, MS-COCO and VOC2007, showcase excellent classification results, highlighting the strong generalization ability of our algorithm. [ABSTRACT FROM AUTHOR]

The task of pedestrian trajectory prediction remains challenging due to variable scenarios, complex social interactions, and uncertainty in pedestrian motion. Previous trajectory prediction research only models from the time domain, which makes it difficult to accurately capture the global and detailed features of complex pedestrian social interactions and the uncertainty of pedestrian movement. These methods also ignore the relationship between scene features and the potential motion patterns of pedestrians. Therefore, we propose a wavelet transform graph convolution network to obtain accurate pedestrian potential motion patterns through time-frequency analysis. We first construct spatial and temporal graphs, then obtain the attention score matrices through the self-attention mechanism in the time domain and combine them with the scene features. Then, we utilize the two-dimensional discrete wavelet transform to generate low-frequency and high-frequency components for representing global and detailed features of spatial-temporal interactions. These components are then further processed using asymmetric convolution, and the wavelet transform adjacency matrix is obtained through the inverse wavelet transform. We then employ graph convolution to combine the graph and the adjacency matrix to obtain spatial and temporal interaction features. Finally, we design the wavelet transform temporal convolution network to directly predict the two-dimensional Gaussian distribution parameters of the future trajectory. Extensive experiments on the ETH, UCY, and SDD datasets demonstrate that our method outperforms the state-of-the-art methods in prediction performance. [ABSTRACT FROM AUTHOR]

Audio descriptions present a tool that helps blind audience members assist theater performances by conveying visual information, such as actors' gestures. However, its high production process cost and effort limit its availability. To address this, we propose a computer vision based system for automated actor gestures recognition, using the state-of-the-art spatio-temporal graph convolution networks (ST-GCNs) for skeleton-based action recognition via transfer learning technique. Hence, we evaluated the transferability of three pre-trained ST-GCNs: the first proposed spatio-temporal graph convolution network (ST-GCN), convolution network of two-stream adaptive graphs (2s-AGCN), and the multi-scale disentangled unified graph convolution network (MS-G3D). We used NTU-RGBD action benchmark as the source domain and collected a novel dataset: TS-RGBD, to serve as the target domain. We then proposed two configurations to accommodate the diversity between the source and target domains. Results showed that ST-GCNs exhibit positive transferability enhancing the models' recognition performance in theatre contexts, promoting automated system for gesture accessibility in theaters. [ABSTRACT FROM AUTHOR]

Studying gene regulatory networks (GRNs) is paramount for unraveling the complexities of biological processes and their associated disorders, such as diabetes, cancer, and Alzheimer’s disease. Recent advancements in computational biology have aimed to enhance the inference of GRNs from gene expression data, a non-trivial task given the networks’ intricate nature. The challenge lies in accurately identifying the myriad interactions among transcription factors and target genes, which govern cellular functions. This research introduces a cutting-edge technique, EGRC (Effective GRN Inference applying Graph Convolution with Self-Attention Graph Pooling), which innovatively conceptualizes GRN reconstruction as a graph classification problem, where the task is to discern the links within subgraphs that encapsulate pairs of nodes. By leveraging Spearman’s correlation, we generate potential subgraphs that bring nonlinear associations between transcription factors and their targets to light. We use mutual information to enhance this, capturing a broader spectrum of gene interactions. Our methodology bifurcates these subgraphs into ‘Positive’ and ‘Negative’ categories. ‘Positive’ subgraphs are those where a transcription factor and its target gene are connected, including interactions among their neighbors. ‘Negative’ subgraphs, conversely, denote pairs without a direct connection. EGRC utilizes dual graph convolution network (GCN) models that exploit node attributes from gene expression profiles and graph embedding techniques to classify these. The performance of EGRC is substantiated by comprehensive evaluations using the DREAM5 datasets. Notably, EGRC attained an AUROC of 0.856 and an AUPR of 0.841 on the E. coli dataset. In contrast, the in silico dataset achieved an AUROC of 0.5058 and an AUPR of 0.958. Furthermore, on the S. cerevisiae dataset, EGRC recorded an AUROC of 0.823 and an AUPR of 0.822. These results underscore the robustness of EGRC in accurately inferring GRNs across various organisms. The advanced performance of EGRC represents a substantial advancement in the field, promising to deepen our comprehension of the intricate biological processes and their implications in both health and disease.

To solve the spectrum scarcity problem, dynamic spectrum access (DSA) technology has emerged as a promising solution. Effectively implementing DSA demands accurate and efficient spectrum prediction. However, complex spatiotemporal correlation and heterogeneity in spectrum observations usually make spectral prediction arduous and even ambiguous. In this letter, we propose a spectrum prediction method based on an attention-aided graph convolutional neural network (AttGCN) to capture features in both spatial and temporal dimensions. By leveraging the attention mechanism, the AttGCN adapts its attention weights at different time steps and spatial positions, thus enabling itself to seize changes in spatiotemporal correlations dynamically. Simulation results show that the proposed spectrum prediction method performs better than baseline algorithms in long-term forecasting tasks.

Abstract In the field of autonomous driving, trajectory prediction of traffic agents is an important and challenging problem. Fully capturing the complex spatio-temporal features in trajectory data is crucial for accurate trajectory prediction. This paper proposes a trajectory prediction model called multi-dimensional spatio-temporal feature fusion (MDSTF), which integrates multi-dimensional spatio-temporal features to model the trajectory information of traffic agents. In the spatial dimension, we employ graph convolutional networks (GCN) to capture the local spatial features of traffic agents, spatial attention mechanism to capture the global spatial features, and LSTM combined with spatial attention to capture the full-process spatial features of traffic agents. Subsequently, these three spatial features are fused using a gate fusion mechanism. Moreover, during the modeling of the full-process spatial features, LSTM is capable of capturing short-term temporal dependencies in the trajectory information of traffic agents. In the temporal dimension, we utilize a Transformer-based encoder to extract long-term temporal dependencies in the trajectory information of traffic agents, which are then fused with the short-term temporal dependencies captured by LSTM. Finally, we employ two temporal convolutional networks (TCN) to predict trajectories based on the fused spatio-temporal features. Experimental results on the ApolloScape trajectory dataset demonstrate that our proposed method outperforms state-of-the-art methods in terms of weighted sum of average displacement error (WSADE) and weighted sum of final displacement error (WSFDE) metrics. Compared to the best baseline model (S2TNet), our method achieves reductions of 4.37% and 6.23% respectively in these metrics.

In the rapidly evolving domain of edge computing, efficient task scheduling emerges as a pivotal challenge due to the increasing complexity and volume of tasks. This study introduces a sophisticated dual-layer hybrid scheduling model that harnesses the strengths of Graph Neural Networks and Deep Reinforcement Learning to enhance the scheduling process. By simplifying task dependencies with Graph Neural Network at the upper layer and integrating Deep Reinforcement Learning with heuristic algorithms at the lower layer, this model optimally allocates tasks, significantly improving scheduling efficiency and reducing response times, particularly beneficial for logistics cloud robots operating in edge computing contexts. We validated the effectiveness of this innovative model through rigorous simulation experiments on the EdgeCloudSim platform, comparing its performance against traditional heuristic methods such as Shortest Job First, First Come First Serve and Heterogeneous Earliest Finish Time. The results confirm that our model consistently achieves superior task scheduling performance across various task volumes, effectively meeting the scheduling demands. This study demonstrates the effectiveness of integrating advanced machine learning techniques with heuristic algorithms to enhance task scheduling processes, making it particularly suitable for scenarios with high demands on response times. This approach not only facilitates more efficient task management but also aligns with the needs of modern edge computing applications, streamlining operations and boosting overall system performance. [ABSTRACT FROM AUTHOR]

Electroencephalogram (EEG) and functional near-infrared spectroscopy (fNIRS) can objectively reflect a person's emotional state and have been widely studied in emotion recognition. However, the effective feature fusion and discriminative feature learning from EEG–fNIRS data is challenging. In order to improve the accuracy of emotion recognition, a graph convolution and capsule attention network model (GCN-CA-CapsNet) is proposed. Firstly, EEG–fNIRS signals are collected from 50 subjects induced by emotional video clips. And then, the features of the EEG and fNIRS are extracted; the EEG–fNIRS features are fused to generate higher-quality primary capsules by graph convolution with the Pearson correlation adjacency matrix. Finally, the capsule attention module is introduced to assign different weights to the primary capsules, and higher-quality primary capsules are selected to generate better classification capsules in the dynamic routing mechanism. We validate the efficacy of the proposed method on our emotional EEG–fNIRS dataset with an ablation study. Extensive experiments demonstrate that the proposed GCN-CA-CapsNet method achieves a more satisfactory performance against the state-of-the-art methods, and the average accuracy can increase by 3–11%. [ABSTRACT FROM AUTHOR]

With the development of precision sensing instruments and data storage devices, the fusion of multi-sensor data in gearbox fault diagnosis has attracted much attention. However, existing methods have difficulty in capturing the local temporal dependencies of multi-sensor monitoring information, and the inescapable noise severely decreases the accuracy of multi-sensor information fusion diagnosis. To address these issues, this paper proposes a fault diagnosis method based on dynamic graph convolutional neural networks and hard threshold denoising. Firstly, considering that the relationships between monitoring data from different sensors change over time, a dynamic graph structure is adopted to model the temporal dependencies of multi-sensor data, and, further, a graph convolutional neural network is constructed to achieve the interaction and feature extraction of temporal information from multi-sensor data. Secondly, to avoid the influence of noise in practical engineering, a hard threshold denoising strategy is designed, and a learnable hard threshold denoising layer is embedded into the graph neural network. Experimental fault datasets from two typical gearbox fault test benches under environmental noise are used to verify the effectiveness of the proposed method in gearbox fault diagnosis. The experimental results show that the proposed DDGCN method achieves an average diagnostic accuracy of up to 99.7% under different levels of environmental noise, demonstrating good noise resistance. [ABSTRACT FROM AUTHOR]

This study introduces a novel pruning approach based on explainable graph convolutional networks, strategically amalgamating pruning and quantisation, aimed to tackle the complexities associated with existing machine learning and deep learning models for stress detection using ultra-short heart rate variability analysis. These complexities often impede the implementation ability of such models on resource-limited devices. The proposed method exhibits exceptional performance, demonstrating high accuracy (97.75%) and efficiency (97.66%) on the WESAD dataset, along with an impressive accuracy (94.48%) and efficiency (94.39%) on the SWELL dataset. Importantly, the runtime complexity saw a significant reduction, down by 63.4% and 69.34% compared to the original model. The proposed method's notable advantage lies in its ability to retain nearly all of the initial model's performance with negligible loss, even when the pruning levels are below 60%. This innovative approach, thus, offers a promising solution for effective stress detection, specifically designed to operate smoothly on devices with limited resources. [ABSTRACT FROM AUTHOR]

As a highly promising technique in the field of similarity search, the hashing-based image retrieval algorithm has received continued attention in recent years because of its strong ability to efficiently provide accurate results when measuring similarities between data instances in the binary space. At the level of practical applications, it is necessary to consider both the optimization of the hashing algorithm itself and the mining of image semantic features to produce high-quality hash codes with a strong semantic resolution, enabling the performance maximisation of such a neighbour search system on complex scenario images. To this end, a unified deep hashing framework termed Dual Enhanced Semantic Hashing (DESH) is proposed in this work. Specifically, it benefits the fast image retrieval in two aspects: 1) By taking advantage of dynamic multi-scale fusion and graph encoding networks, the dual enhanced feature learning significantly strengthens the semantic feature representation by jointly exploring and encoding the local multi-scale information with the high-order adjacency relationship between original images; 2) With the joint optimization of diverse loss functions, the binary semantic modelling process seamlessly module the image semantic information within the hash function learning in the code generation, aiming to generate discriminative hash codes to refine the retrieval performance eventually. By conducting extensive experiments on public datasets, the retrieval results further validate the claims of the proposed DESH by exhibiting its superior performance against competitive state-of-the-art baselines. [ABSTRACT FROM AUTHOR]

Graph neural networks (GNNs) have recently garnered significant attention due to their exceptional performance across various applications, including hyperspectral (HS) image classification. However, most existing GNN-based models for HS image classification are limited depth models and often suffer from performance degradation as model depth increases. This study introduces HyperGCN, an exclusive GNN-based model designed with multiple graph convolutional layers to exploit the rich spectral information inherent in HS images, thereby enhancing classification performance. To address performance degradation, HyperGCN incorporates techniques resistant to oversmoothing into its architecture. Additionally, multiple-side exit branches are integrated into the intermediate layers of HyperGCN, enabling dynamic management of the complexity of HS images. Less complex HS images are processed by fewer layers, exiting early via attached branches, while more complex images traverse multiple layers until reaching the final output layer. Extensive experiments on four benchmark HS datasets (Indian Pines, Pavia University, Salinas, and Botswana) demonstrate HyperGCN's superior performance over basic GNN-based models. Notably, HyperGCN outperforms or performs comparably to the CNN-GNN combined model in classifying HS images. Furthermore, the superior performance of multi-exit HyperGCN over its single-exit counterpart emphasizes the effectiveness of incorporating side exit branches in GNN-based HS image classification. Compared to state-of-the-art models, multi-exit HyperGCN demonstrates competitive performance, highlighting its effectiveness in handling complex spectral information in HS images while maintaining an acceptable balance between accuracy and computational efficiency. [ABSTRACT FROM AUTHOR]

Semi-supervised graph convolutional networks (SSGCNs) have been proven to be effective in hyperspectral image classification (HSIC). However, limited training data and spectral uncertainty restrict the classification performance, and the computational demands of a graph convolution network (GCN) present challenges for real-time applications. To overcome these issues, a dual-branch fusion of a GCN and convolutional neural network (DFGCN) is proposed for HSIC tasks. The GCN branch uses an adaptive multi-scale superpixel segmentation method to build fusion adjacency matrices at various scales, which improves the graph convolution efficiency and node representations. Additionally, a spectral feature enhancement module (SFEM) enhances the transmission of crucial channel information between the two graph convolutions. Meanwhile, the CNN branch uses a convolutional network with an attention mechanism to focus on detailed features of local areas. By combining the multi-scale superpixel features from the GCN branch and the local pixel features from the CNN branch, this method leverages complementary features to fully learn rich spatial–spectral information. Our experimental results demonstrate that the proposed method outperforms existing advanced approaches in terms of classification efficiency and accuracy across three benchmark data sets. [ABSTRACT FROM AUTHOR]

In recent years, self-supervised representation learning for skeleton-based action recognition has achieved remarkable results using skeleton sequences with the advance of contrastive learning methods. However, existing methods often overlook the local information within the skeleton data, so as to not efficiently learn fine-grained features. To leverage local features to enhance representation capacity and capture discriminative representations, we design an adaptive self-supervised contrastive learning framework for action recognition called AdaSCLR. In AdaSCLR, we introduce an adaptive spatiotemporal graph convolutional network to learn the topology of different samples and hierarchical levels and apply an attention mask module to extract salient and non-salient local features from the global features, emphasizing their significance and facilitating similarity-based learning. In addition, AdaSCLR extracts information from the upper and lower limbs as local features to assist the model in learning more discriminative representation. Experimental results show that our approach is better than the state-of-the-art methods on NTURGB+D, NTU120-RGB+D, and PKU-MMD datasets. [ABSTRACT FROM AUTHOR]

Multi-interest sequential recommendations leverage users' historical behavior to provide recommendations that match multiple interests. Most of these methods have not fully extracted higher-order information hidden in users' interactions and have overlooked the multiple features of items. To this end, this paper proposes a multi-interest model called "multi-interest sequential recommendation with simplified graph convolution and item multi-features (SGCMF)". Firstly, a simplified graph convolution module is designed based on bipartite graphs, which utilizes mean pooling to aggregate neighboring information and employs a feedforward neural network (FNN) for nonlinear transformations and combinations. This method reduces redundant information and captures higher-order relationships, thereby simplifying the complexity of modeling high-order interactions and improving prediction accuracy. Secondly, an item multi-feature extraction module is proposed, which represents item features with multiple vectors, and analyzes each feature from multiple perspectives while preserving important relationships between features. The model correlates multiple features of the item with user interests, thereby achieving a fine-grained analysis of user interests. Extensive experiments are conducted on five real-world scenarios, and the results are compared with state-of-the-art methods. The experimental results show that SGCMF outperforms other baselines. [ABSTRACT FROM AUTHOR]

Recommendation systems research is a subfield of information retrieval, as these systems recommend appropriate items to users during their visits. Appropriate recommendation results will help users save time searching while increasing productivity at work, travel, or shopping. The problem becomes more difficult when the items are geographical locations on the ground, as they are associated with a wealth of contextual information, such as geographical location, opening time, and sequence of related locations. Furthermore, on social networking platforms that allow users to check in or express interest when visiting a specific location, their friends receive this signal by spreading the word on that online social network. Consideration should be given to relationship data extracted from online social networking platforms, as well as their impact on the geolocation recommendation process. In this study, we compare the similarity of geographic locations based on their distance on the ground and their correlation with users who have checked in at those locations. When calculating feature embeddings for users and locations, social relationships are also considered as attention signals. The similarity value between location and correlation between users will be exploited in the overall architecture of the recommendation model, which will employ graph convolution networks to generate recommendations with high precision and recall. The proposed model is implemented and executed on popular datasets, then compared to baseline models to assess its overall effectiveness. [ABSTRACT FROM AUTHOR]

Spatio-temporal feature extraction and fusion are crucial to traffic prediction accuracy. However, the complicated spatio-temporal correlations and dependencies between traffic nodes make the problem quite challenging. In this paper, a multi-scale spatio-temporal network (MSSTN) is proposed to exploit complicated local and nonlocal correlations in traffic flow for traffic prediction. In the proposed method, a convolutional neural network, a self-attention module, and a graph convolution network (GCN) are integrated to extract and fuse multi-scale temporal and spatial features to make predictions. Specifically, a self-adaption temporal convolutional neural network (SATCN) is first employed to extract local temporal correlations between adjacent time slices. Furthermore, a self-attention module is applied to capture the long-range nonlocal traffic dependence in the temporal dimension and fuse it with the local features. Then, a graph convolutional network module is utilized to learn spatio-temporal features of the traffic flow to exploit the mutual dependencies between traffic nodes. Experimental results on public traffic datasets demonstrate the superiority of our method over compared state-of-the-art methods. The ablation experiments confirm the effectiveness of each component of the proposed model. Our implementation on Pytorch is publicly available at https://github.com/csust-sonie/MSSTN. [ABSTRACT FROM AUTHOR]

Piwi-interacting RNA (piRNA) is a type of non-coding small RNA that is highly expressed in mammalian testis. PiRNA has been implicated in various human diseases, but the experimental validation of piRNA-disease associations is costly and time-consuming. In this article, a novel computational method for predicting piRNA-disease associations using a multi-channel graph variational autoencoder (MC-GVAE) is proposed. This method integrates four types of similarity networks for piRNAs and diseases, which are derived from piRNA sequences, disease semantics, piRNA Gaussian Interaction Profile (GIP) kernel, and disease GIP kernel, respectively. These networks are modeled by a graph VAE framework, which can learn low-dimensional and informative feature representations for piRNAs and diseases. Then, a multi-channel method is used to fuse the feature representations from different networks. Finally, a three-layer neural network classifier is applied to predict the potential associations between piRNAs and diseases. The method was evaluated on a benchmark dataset containing 5,002 experimentally validated associations with 4,350 piRNAs and 21 diseases, constructed from the piRDisease v1.0 database. It achieved state-of-the-art performance, with an average AUC value of 0.9310 and an AUPR value of 0.9247 under five-fold cross-validation. This demonstrates the method's effectiveness and superiority in piRNA-disease association prediction. [ABSTRACT FROM AUTHOR]

In the modern digital age, users are exposed to a vast amount of content and information, and the importance of recommendation systems is increasing accordingly. Traditional recommendation systems mainly use matrix factorization and collaborative filtering methods, but problems with scalability due to an increase in the amount of data and slow learning and inference speeds occur due to an increase in the amount of computation. To overcome these problems, this study focused on optimizing LightGCN, the basic structure of the graph-convolution-network-based recommendation system. To improve this, techniques and structures were proposed. We propose an embedding enhancement method to strengthen the robustness of embedding and a non-combination structure to overcome LightGCN's weight sum structure through this method. To verify the proposed method, we have demonstrated its effectiveness through experiments using the SELFRec library on various datasets, such as Yelp2018, MovieLens-1M, FilmTrust, and Douban-book. Mainly, significant performance improvements were observed in key indicators, such as Precision, Recall, NDCG, and Hit Ratio in Yelp2018 and Douban-book datasets. These results suggest that the proposed methods effectively improved the recommendation performance and learning efficiency of the LightGCN model, and the improvement of LightGCN, which is most widely used as a backbone network, makes an important contribution to the entire field of GCN-based recommendation systems. Therefore, in this study, we improved the learning method of the existing LightGCN and changed the weight sum structure to surpass the existing accuracy. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Tiangong University is the property of Journal of Tianjin Polytechnic University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Targeting the challenge of variable working conditions in bearing fault diagnosis, most of the fault diagnosis methods based on transfer learning focus on the transfer of knowledge, resulting in a poor diagnosis effect in the target domain. To solve the problem of transfer performance degradation, a multi-perception graph convolution transfer network (MPGCTN) is proposed. The MPGCTN is composed of a graph generation module, graph perception module, and domain discrimination module. In the graph generation module, a one-dimensional convolution neural network (1-D CNN) is used to extract features from the input, and then the structural features of samples are mined in the graph generation layer to construct the sample graph. In the following graph perception module, a multi-perception graph convolution network is designed to model the sample graph and learn the data structure information of the sample. Finally, in the domain discrimination module, the method is used to align the structural differences of the case graphs in different domains. Experimental results from experiments on Case Western Reserve University (CWRU) and Paderborn University (PU) bearing datasets show that the proposed method is effective and superior. [ABSTRACT FROM AUTHOR]

The relationship between users and items, which cannot be recovered by traditional techniques, can be extracted by the recommendation algorithm based on the graph convolution network. The current simple linear combination of these algorithms may not be sufficient to extract the complex structure of user interaction data. This paper presents a new approach to address such issues, utilizing the graph convolution network to extract association relations. The proposed approach mainly includes three modules: Embedding layer, forward propagation layer, and score prediction layer. The embedding layer models users and items according to their interaction information and generates initial feature vectors as input for the forward propagation layer. The forward propagation layer designs two parallel graph convolution networks with self-connections, which extract higher-order association relevance from users and items separately by multi-layer graph convolution. Furthermore, the forward propagation layer integrates the attention factor to assign different weights among the hop neighbors of the graph convolution network fusion, capturing more comprehensive association relevance between users and items as input for the score prediction layer. The score prediction layer introduces MLP (multi-layer perceptron) to conduct nonlinear feature interaction between users and items, respectively. Finally, the prediction score of users to items is obtained. The recall rate and normalized discounted cumulative gain were used as evaluation indexes. The proposed approach effectively integrates higher-order information in user entries, and experimental analysis demonstrates its superiority over the existing algorithms. [ABSTRACT FROM AUTHOR]

Traffic flow prediction is considered to be one of the fundamental technologies in intelligent transportation systems (ITSs) with a tremendous application prospect. Unlike traditional time series analysis tasks, the key challenge in traffic flow prediction lies in effectively modelling the highly complex and dynamic spatiotemporal dependencies within the traffic data. In recent years, researchers have proposed various methods to enhance the accuracy of traffic flow prediction, but certain issues still persist. For instance, some methods rely on specific static assumptions, failing to adequately simulate the dynamic changes in the data, thus limiting their modelling capacity. On the other hand, some approaches inadequately capture the spatiotemporal dependencies, resulting in the omission of crucial information and leading to unsatisfactory prediction outcomes. To address these challenges, this paper proposes a model called the Dynamic Spatial–Temporal Self-Attention Network (DSTSAN). Firstly, this research enhances the interaction between different dimension features in the traffic data through a feature augmentation module, thereby improving the model's representational capacity. Subsequently, the current investigation introduces two masking matrices: one captures local spatial dependencies and the other captures global spatial dependencies, based on the spatial self-attention module. Finally, the methodology employs a temporal self-attention module to capture and integrate the dynamic temporal dependencies of traffic data. We designed experiments using historical data from the previous hour to predict traffic flow conditions in the hour ahead, and the experiments were extensively compared to the DSTSAN model, with 11 baseline methods using four real-world datasets. The results demonstrate the effectiveness and superiority of the proposed approach. [ABSTRACT FROM AUTHOR]

Brain-computer interface (BCI) technology based on motor imagery (MI) EEG signals has been widely concerned and studied in the medical application of motor function rehabilitation for stroke patients. However, the MI signal has the characteristics of low signal-to-noise ratio and large volume variability, which leads to excessive noise in the EEG signal and affects the classification performance. Therefore, how to fully extract MI signal features to obtain higher singlesubject classification accuracy, and how to train a general model with excellent cross-subject performance are urgent problems to be solved when MI-BCI system is used in practical applications. In response to this problem, this paper proposes a spatiotemporal adaptive graph convolutional network model for different subjects, which extracts MI feature signals from two dimensions of time and spatio for classification. The model includes four modules: spatial adaptive graph convolution module, temporal adaptive graph convolution module, feature fusion module and feature classification module. The spatial adaptive graph convolution module dynamically constructs the spatial graph representation through feature similarity between channels, and gets rid of the limitation of artificially constructs graph representation. The time- adaptive graph convolution module divides the time series of EEG signals into multiple time segments and calculates the similarity between time segments, so as to adaptively construct the time map representation of EEG signals and eliminate the influence of noise. Finally, feature fusion and classification are performed. The results show that the proposes method achieves an average classification accuracy of 90.45% and 91.64% is achieved by using 10-fold cross-validation method on BCIIV2a dataset and 91.64% on HGD dataset. Compared with the current state-of-the-art methods, this method achieves a higher accuracy rate, proving the effectiveness of our model. By using transfer learning to experiment on different individuals, the average accuracy is increased by 1.66 percentage points, which proves the robustness of the model. [ABSTRACT FROM AUTHOR]

Aiming at the problem that existing image text cross-modal retrieval is difficult to fully exploit the local consistency in the mode in the common subspace, a cross-modal retrieval method based on improved graph convolution is proposed. In order to improve the local consistency within each mode, the modal diagram is constructed with a single sample as a node, fully mining the interactive information between features. In order to solve the problem that graph convolution network can only do shallow learning, the method of adding initial residual link and weight identity map in each layer of graph convolution is adopted to alleviate this phenomenon. In order to jointly update the central node features through higher-order and lower-order neighbor information, an improvement is proposed to reduce neighbor nodes and increase the number of layers in graph convolution network. In order to learn highly locally consistent and semantically consistent public representation, it shares the weights of common representation learning layer, and jointly optimizes the semantic constraints within the modes and the modal invariant constraints between modes in the common subspace. The experimental results show that on the two cross-modal data sets of Wikipedia and Pascal sentence, the average mAP values of different retrieval tasks are 2.2%~42.1% and 3.0%~54.0% higher than the 11 existing methods. [ABSTRACT FROM AUTHOR]