Search

• Multi-pattern connection graph convolutional networks can effectively identify essential tremor with resting tremor and tremor-dominant Parkinson's disease. • Different connection modes may provide distinct discriminative information for diagnosis. • The occipital network and basal ganglion-temporal lobe networks appear to be tremor-related networks for rET and tPD, respectively. Essential tremor with resting tremor (rET) and tremor-dominant Parkinson's disease (tPD) share many similar clinical symptoms, leading to frequent misdiagnoses. Functional connectivity (FC) matrix analysis derived from resting-state functional MRI (Rs-fMRI) offers a promising approach for early diagnosis and for exploring FC network pathogenesis in rET and tPD. However, methods relying solely on a single connection pattern may overlook the complementary roles of different connectivity patterns, resulting in reduced diagnostic differentiation. Therefore, we propose a multi-pattern connection Graph Convolutional Network (MCGCN) method to integrate information from various connection modes, distinguishing between rET and healthy controls (HC), tPD and HC, and rET and tPD. We constructed FC matrices using three different connectivity modes for each subject and used these as inputs to the MCGCN model for disease classification. The classification performance of the model was evaluated for each connectivity mode. Subsequently, gradient-weighted class activation mapping (Grad-CAM) was used to identify the most discriminative brain regions. The important brain regions identified were primarily distributed within cerebellar-motor and non-motor cortical networks. Compared with single-pattern GCN, our proposed MCGCN model demonstrated superior classification accuracy, underscoring the advantages of integrating multiple connectivity modes. Specifically, the model achieved an average accuracy of 88.0% for distinguishing rET from HC, 88.8% for rET from tPD, and 89.6% for tPD from HC. Our findings indicate that combining graph convolutional networks with multi-connection patterns can not only effectively discriminate between tPD, rET, and HC but also enhance our understanding of the functional network mechanisms underlying rET and tPD. [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]

Social media has evolved into a widely influential information source in contemporary society. However, the widespread use of social media also enables the rapid spread of fake news, which can pose a significant threat to national and social stability. Current fake news detection methods primarily rely on graph neural network, which analyze the dissemination patterns of news articles. Nevertheless, these approaches frequently overlook the semantic characteristics of the news content itself. To address this problem, we propose a novel Graph Convolution Network and BERT combined with Co-Attention (GBCA) model. Initially, we conduct training for a graph classification task on the propagation structure of fake news. Subsequently, we employ the BERT model to extract semantic features in fake news. Finally, we utilize co-attention mechanism to integrate the two dimensions of propagation structure and semantic features, which enhances the effectiveness of fake news detection. Our model outperforms baseline methods in terms of accuracy and training time, as demonstrated by experiments on three public benchmark datasets. • We propose a novel fake news detection model that combines the graph convolution network and BERT. • We employ the co-attention mechanism to obtain better results and achieve training convergence in a shorter time. • We conduct ablation studies and result analysis to understand how to work of our model and where the performance gain comes from. [ABSTRACT FROM AUTHOR]

Graph convolutional network has been extensively employed in semi-supervised classification tasks. Although some studies have attempted to leverage graph convolutional networks to explore multi-view data, they mostly consider the fusion of feature and topology individually, leading to the underutilization of the consistency and complementarity of multi-view data. In this paper, we propose an end-to-end joint fusion framework that aims to simultaneously conduct a consistent feature integration and an adaptive topology adjustment. Specifically, to capture the feature consistency, we construct a deep matrix decomposition module, which maps data from different views onto a feature space obtaining a consistent feature representation. Moreover, we design a more flexible graph convolution that allows to adaptively learn a more robust topology. A dynamic topology can greatly reduce the influence of unreliable information, which acquires a more adaptive representation. As a result, our method jointly designs an effective feature fusion module and a topology adjustment module, and lets these two modules mutually enhance each other. It takes full advantage of the consistency and complementarity to better capture the more intrinsic information. The experimental results indicate that our method surpasses state-of-the-art semi-supervised classification methods. • Propose an end-to-end framework for multi-view semi-supervised classification. • Design a multi-view auto-encoder to fuse feature by approximating the matrix decomposition. • Explore a more robust topology that fuses the adjacency matrices generated by k NN and k FN. [ABSTRACT FROM AUTHOR]

In recent years, the application of convolutional neural networks (CNNs) and graph convolutional networks (GCNs) in hyperspectral image classification (HSIC) has achieved remarkable results. However, the limited label samples are still a major challenge when using CNN and GCN to classify hyperspectral images. In order to alleviate this problem, a double branch fusion network of CNN and enhanced graph attention network (CEGAT) based on key sample selection strategy is proposed. First, a linear discrimination of spectral inter-class slices (LD_SICS) module is designed to eliminate spectral redundancy of HSIs. Then, a spatial spectral correlation attention (SSCA) module is proposed, which can extract and assign attention weight to the spatial and spectral correlation features. On the graph attention (GAT) branch, the HSI is segmented into some super pixels as input to reduce the amount of network parameters. In addition, an enhanced graph attention (EGAT) module is constructed to enhance the relationship between nodes. Finally, a key sample selection (KSS) strategy is proposed to enable the network to achieve better classification performance with few labeled samples. Compared with other state-of-the-art methods, CEGAT has better classification performance under limited label samples. [ABSTRACT FROM AUTHOR]

In the 3D skeleton-based action recognition task, learning rich spatial and temporal motion patterns from body joints are two foundational yet under-explored problems. In this paper, we propose two methods for improving these problems: (I) a novel glimpse-focus action recognition strategy that captures multi-range pose features from the whole body and key body parts jointly; (II) a powerful temporal feature extractor JD-TC that enriches trajectory features by inferring different inter-frame correlations for different joints. By coupling these two proposals, we develop a powerful skeleton-based action recognition system that extracts rich pose and trajectory features from a skeleton sequence and outperforms previous state-of-the-art methods on three large-scale datasets. [ABSTRACT FROM AUTHOR]

• A few-shot model HDGN is proposed to improve the acquisition of on-graph filtering information in the spectral domain. • A mixed metric combining nonlinear and linear metrics is designed to update edge features more robustly. • HDGN enhances the acquisition of low-frequency information on the graph through the heat kernel function. • The low-frequency information learned from the graph can achieve better performance in few-shot classification tasks. A heat diffusion graph network (HDGN) is proposed in this paper, which retains more similar graph signals in the spectral domain, for few-shot learning. Convolution on the graph is essentially the filtering of the graph signal. Most existing graph-network-based few-shot learning methods process graph signals with high-pass filters to get the difference in information. However, the low-frequency similar information is usually more valuable in the few-shot tasks. A joint low-pass filter is constructed to filter low-frequency graph signals, that is, heat kernel convolution aggregates similar information from neighboring nodes. The obtained low-frequency similarity information is utilized to update the representations of nodes on the graph. In addition, a more robust mixed metric is designed to dynamically update the edge feature of the graph. Predicting Unknown Nodes on Graphs by Alternating Updates of Node Representation and Edge Matrix. The experimental results also demonstrate that HDGN achieves better performance for the few-shot classification task. [ABSTRACT FROM AUTHOR]

• This is a comprehensive review of Transformer for Skeleton-based Action Recognition. • This paper proposes a new taxonomy of transformer-style techniques. • This survey aims to help researchers systematically select promising future directions. Skeleton-based action recognition has rapidly become one of the most popular and essential research topics in computer vision. The task is to analyze the characteristics of human joints and accurately classify their behaviors through deep learning technology. Skeleton provides numerous unique advantages over other data modalities, such as robustness, compactness, noise immunity, etc. In particular, the skeleton modality is extremely lightweight, which is especially beneficial for deep learning research in low-resource environments. Due to the non-European nature of skeleton data, Graph Convolution Network (GCN) has become mainstream in the past few years, leveraging the benefits of processing topological information. However, with the explosive development of transformer methods in natural language processing and computer vision, many works have applied transformer into the field of skeleton action recognition, breaking the accuracy monopoly of GCN. Therefore, we conduct a survey using transformer method for skeleton-based action recognition, forming of a taxonomy on existing works. This paper gives a comprehensive overview of the recent transformer techniques for skeleton action recognition, proposes a taxonomy of transformer-style techniques for action recognition, conducts a detailed study on benchmark datasets, compares the algorithm accuracy of standard methods, and finally discusses the future research directions and trends. To the best of our knowledge, this study is the first to describe skeleton-based action recognition techniques in the style of transformers and to suggest novel recognition taxonomies in a review. We are confident that Transformer-based action recognition technology will become mainstream in the near future, so this survey aims to help researchers systematically learn core tasks, select appropriate datasets, understand current challenges, and select promising future directions. [ABSTRACT FROM AUTHOR]

• We propose a new graph convolution network with subgraph embedding. • It extracts local information with subgraphs partitioned by Girvan Newman algorithm. • Experiments confirm the superiority of predicting mutagenicity in aromatic hydrocarbons. • The accuracy is verified with comparison of nine SOTA deep learning models. • The proposed method prevents about 15 %p of GCN's information dilution. An aromatic hydrocarbon refers to an organic material having a carbon ring such as benzene and a functional group in the carbon ring. As the industry develops, natural pollution becomes harsh, new compounds emerge, and the exposure to aromatic hydrocarbons is continuously increasing. Predicting mutagenicity is one of the crucial issues in reducing the risk because these organisms may have properties that penetrate the DNA of living things to cause mutations. Recently, the accuracy of mutation prediction has improved due to the power of deep learning. However, most conventional methods do not consider the characteristics of molecular aromatic hydrocarbons, which dilutes local information and results in a severe deterioration of the prediction performance. In this paper, we propose a method of exploiting subgraph convolution neural networks that enables the extraction of local information of a graph by partitioning it to maintain the detailed information. For extracting the features of molecules, we use the Girvan Newman algorithm to partition the graph according to the carbon ring and functional group and obtain the embedding vectors of the subgraphs as well as the original graph with graph convolution network (GCN). The embedding vectors are combined to represent the whole graph information and predict mutagenicity. Experiments with MUTAG, NCI1 and NCI109, datasets for predicting mutagenicity of molecules in graph structure, confirm that we successfully segment carbon rings and functional groups from molecular graphs and predict mutations using the partitioned graphs, leading to a 2 %p performance improvement. In addition, the proposed method has prevented about 15 %p of information dilution in GCN, and an analysis of the latent space of graphs reveals that the subgraphs extracted maintain the local information appropriately. [ABSTRACT FROM AUTHOR]

• A DFGCN of transmission system is proposed from a multi-sensor-multi-condition view. • Multi-branch parallel GCN realizes the internal correlation learning. • MR-Softmax is designed to obtain multi-representation characteristics. • The entropy weighting is used to improve the evidence theory. Power transmission reliability of drivelines guarantees fast maneuverability of heavy vehicles. During health monitoring, multi-sensor data fusion technology has been widely used in the improvement of fault diagnosis accuracy of long-link multi-structure drivelines. However, in multi-sensor fusion and joint fault diagnosis scenarios where multiple conditions coexist, it is still challenging to fuse multi-sensor data and extract generalized fault intrinsic features for any combination of operating conditions under multi-sensor monitoring. In this paper, a dual fusion graph convolutional network (DFGCN) is proposed for multi-sensor-multi-condition fault diagnosis of the transmission system. First, considering the data structure and the correlation of different sensors, DFGCN constructs multi-sensor intrinsic links synchronously from the data and feature levels by using multi-branch parallel GCN. Second, considering the susceptibility to over-confidence when the feature space of multi-condition data is inconsistent, an entropy-weighted multi-representation Dempster-Shafer (EWMR-DS) evidence theory fusion strategy is designed to extract the condition-shared features by increasing the label space diversity. Finally, an end-to-end lightweight diagnosis framework is scalable to multi-sensor and multi-working conditions in engineering practice, and the dual information fusion improves the fusion efficiency of fine-grained features with distributional differences. Using experimental datasets collected from two typical transmission fault test benches, the effectiveness of the proposed DFGCN method in multi-sensor-multi-condition scenarios is verified. The results indicate that DFGCN achieves an average diagnostic accuracy of more than 99.7% and superior noise resistance under different degrees of environmental noise. [ABSTRACT FROM AUTHOR]

A continuous real-time video monitoring for preterm infants that suffer from fragile condition attracts increasing attention due to its significance in the early screening of diseases. Specially, the limb movement recognition of preterm infants in neonatal intensive care units (NICUs) plays an important role for assessing the health status of preterm infants. The main challenges of the preterm infant limb movement recognition are arisen from the irregular movement speed and the multi-limb classification. To address these issues, a novel convolution graph recognition network (CGRN) is proposed by integrating three-dimensional (3D) convolution neural networks (CNNs) with graph convolution networks (GCNs). Particularly, the spatial–temporal frequency characteristics of different channels in 3D CNN generally have poor adaptability to characterize fast limb movement. Thus, GCN is merged to enhance the channel connection and improve the capabilities of representation learning. The node filtering and topological connection are learned by GCN to establish dependencies among different channels. A fusion multi-label loss function consisting of the Binary Cross Entropy (BCE) and the Multi-label Soft-Margin (MSM) loss is further developed to train the proposed CGRN. Experiments on the preterm Infant Activity Dataset (IAD) are conducted to demonstrate the effectiveness of the proposed CGRN algorithm. The ablations studies indicate that the structure optimization process and the auxiliary performance of Laplace matrices are worked effectively, and the proposed CGRN obtains the average accuracy of 94.1%, the CF1 value of 88.4%, and the OF1 score of 91.7%. • A novel CGRN is developed for limb movement recognition of premature infants. • The CNN is used to reduce GCN's dependence on formatted data. • The Laplacian matrix is used to represent the relationship of feature channels. • Model fusion addresses CNN channel independence and GCN data restrictions. • Unlike the fully connected layer, the GCN layer uses Laplacian matrix weighting. [ABSTRACT FROM AUTHOR]

The advancement of digital technology has driven innovation across industries. Artificial intelligence (AI) as a pivotal digital technology, has the potential to stimulate innovative outcomes when converged with construction technologies. However, current research has yet to accurately discover promising technology opportunities between AI and construction. To fill this gap, this study uses patent data to develop a holistic approach for precisely discovering technology opportunities. It focuses on analyzing technology convergence from two distinct perspectives: the "degree" and "value" of convergence. The technology convergence degree is measured based on the semantic network and graph convolution network (GCN) method. The technology convergence value is calculated according to patent attribute metrics. Analysis of the spatial-temporal evolution of technology convergence reveals the irreversible convergence trend between AI and construction. This study further identifies key opportunities, such as the convergence of computer vision and construction equipment technology, which have the potential to advance the construction industry. Technology opportunities are also identified both at global and national levels, providing strategic insights that could guide industry regulatory authorities. [Display omitted] • A framework is proposed to identify the technology opportunities from a technology convergence perspective. • The spatio-temporal evolution trend of the convergence between AI and construction technology is revealed. • The key convergence fields of both global and national convergence of AI and construction technologies are discussed. [ABSTRACT FROM AUTHOR]

Recently, there has been a surge of interest in recommendation systems that leverage knowledge graphs, primarily because of their effectiveness in addressing sparsity and cold-start challenges inherent in collaborative filtering approaches. In most previous studies, researchers have focused on the way knowledge associations are encoded in knowledge graphs, but have not sufficiently highlighted the signals of collaboration that are implicit in the interaction between users and items. As a result, the learned embeddings do not provide a complete representation of the semantic information. In this paper, we describe a new model called a heterogeneous propagation graph convolution network for a recommendation system combined with a knowledge graph (HP-GCN). It adopts a heterogeneous propagation to generate user embedding representations, thereby combining encoded collaborative signals and auxiliary knowledge in knowledge graphs. Furthermore, we incorporate an attention mechanism to differentiate the contributions made by diverse neighbors as opposed to those made by users. Since most graph convolutions tend to suffer from over-smoothing when the number of convolutional layers increases, leading to insufficient utilization of high-order information, this paper uses an improved graph convolution strategy to generate item embeddings. This strategy has two different aggregation mechanisms embedding into different subgraphs, which can more fully utilize high-order information and mitigate the over-smoothing problem. Thus, we are able to efficiently prevent negative information originating from higher-order neighbors into the process of embedding learning. In extensive experiments, we applied HP-GCN to four large-scale real datasets for music, books, movies, and restaurants. The experimental outcomes revealed that HP-GCN generally surpassed the baseline methods in both recommendation accuracy and diversity, showing superior recommendation performance overall. • A novel knowledge graph recommendation method is proposed. • Learning ripple set embedding to capture the user's latent interest. • Using step-wise graph convolution to solve the over-smoothing problem. • The experiments on four real datasets demonstrate the effectiveness of our model. [ABSTRACT FROM AUTHOR]

As technology advances, competition among market players intensifies, highlighting the importance of comprehending both one's own and competitors' pricing strategies. Traditional approaches often rely on static factors for price forecasting, disregarding the dynamic nature of market competition. In contrast to methods utilizing a static graph structure and uniform weights for predictions, we introduce STAD-GCN (Spatial–Temporal Attention-based Dynamic Graph Convolutional Network). This innovative model dynamically incorporates competitive factors into price prediction by employing attention mechanisms and graph convolution operations. Such an approach allows for the adaptation of graph structures and node relationships in response to temporal and spatial changes, offering a more nuanced understanding of market dynamics. To test STAD-GCN, we utilized international and domestic oil price data alongside specific gas station details from the South Korea National Oil Corporation, focusing on Seoul and Busan. The model presents remarkable performance, achieving mean absolute errors of 12.648 in Seoul and 11.242 in Busan, surpassing state-of-the-art models. Based on the findings of our research, we present some academic and practical implications, as well as several future research directions. • Referencing competitors' prices is essential due to consumer sensitivity. • STAD-GCN integrates competitive elements and price factors dynamically. • The model improves price prediction via spatial–temporal attention mechanisms. • STAD-GCN consistently maintains low MAE and RMSE in long-term predictions. [ABSTRACT FROM AUTHOR]

Software vulnerability detection is a challenging task in the security field, the boom of deep learning technology promotes the development of automatic vulnerability detection. Compared with sequence-based deep learning models, graph neural network (GNN) can learn the structural features of code, it performs well in the field of vulnerability detection for source code. However, different GNNs have different detection results for the same code, and using a single kind of GNN may lead to high false positive rate and false negative rate. In addition, the complex structure of source code causes single GNN model cannot effectively learn their depth feature, thereby leading to low detection accuracy. To solve these limitations, we propose a software vulnerability detection model called iGnnVD based on the integrated graph neural networks. In the proposed iGnnVD model, the base detectors including GCN, GAT and APPNP are first constructed to capture the bidirectional information in the code graph structure with bidirectional structure; And then, the residual connection is used to aggregate the features while retaining the features each time; Finally, the convolutional layer is used to perform the aggregated classification. In addition, an integration module that analyzes the detection results of three detectors for final classification is designed using a voting strategy to solve the problem of high false positive rate and false negative rate caused by using a single kind of base detector. We perform extensive experiments on three datasets and experimental results show that the proposed iGnnVD model can improve the detection accuracy of vulnerabilities in source code as well as reduce the false positive rate and false negative rate compared with existing deep learning-based vulnerability detection models, it also has good stability. [ABSTRACT FROM AUTHOR]

Structural deep clustering involves the use of neural networks for fusing semantic and structural representations for clustering tasks, and it has been receiving increasing attention. In some pioneering works, auto-encoder (AE)-specific representations were integrated with a graph convolutional network (GCN)-specific representation by delivering semantic information to the GCN module layer-by-layer. Although promising performance has been achieved in various applications, we observed that a vital aspect was overlooked in these works: the structural information may vanish in the learning process because of the over-smoothing problem of the GCN module, leading to non-representative features and, thus, deteriorating clustering performance. In this study, we address this issue by proposing a structure enhanced deep clustering network. The GCN-specific structural data representation is enhanced and supervised by its structural information. Specifically, the GCN-specific structural data representation is strengthened during the learning process by combining it with a structure enhanced semantic (SES) representation. A novel structure enhanced AE, named the weighted neighbourhood AE (wNAE), is employed to learn the SES representation for each data sample. Finally, we design a joint supervision strategy to uniformly guide the simultaneous learning of the wNAE and GCN modules and the clustering assignment. Experimental results for different datasets empirically validate the importance of semantic and neighbour-wise structure learning. [ABSTRACT FROM AUTHOR]

• A center-connected graph convolutional network enhanced with salient image features (SIFE-CGCN) was proposed by integrating the image semantics to improve the recognition performance of similar actions. • A center-connected skeleton topology was developed to enhance the learning capability of the GCN on the potential cooperative dependencies of all joints. • The DTW-based metric was developed to measure the action similarity and build the similar action dataset. The proposed model achieves state-of-the-art performance on three large-scale datasets. Skeleton-based human action recognition has drawn more and more attention due to its easy implementation and stable application in intelligent human-robot interaction. However, most existing studies only used the skeleton data but completely ignored other image semantic information to build action recognition models, which would confuse the recognition of similar actions because of the ambiguity between skeleton data. Here, a center-connected graph convolutional network enhanced with salient image features (SIFE-CGCN) is proposed to address the problem of similar action recognition. First, a center-connected network (CGCN) is constructed to capture the small differences between similar actions through exploring the possible collaboration between all joints. Subsequently, a metric of movement changes is employed to optimally select the salient image from an action video, and then the EfficientNet is used to achieve the action semantic classification of the salient images. Finally, the recognition results of CGCN are strengthened with the classification results of salient images to further improve the recognition accuracy for similar actions. Additionally, a metric is proposed to measure the action similarity with the skeleton data, and then a similar action dataset is built. Extensive experiments on the datasets of similar action and NTU RGB + D 60/120 were conducted to verify the performance of the proposed methods. Experimental results validated the effectiveness of salient image feature enhancement and showed that the proposed SIFE-CGCN achieved the state-of-the-art performance on the similar action and NTU RGB + D 60/120 datasets. [ABSTRACT FROM AUTHOR]

The proliferation of portable electroencephalogram (EEG) recording devices has made it practically feasible to develop the motor imagery (MI) based brain–computer interfaces (BCIs). However, the low signal-to-noise ratio of EEG signals for abstract MI tasks, limited data, limited EEG channels, and strong inter- and intra-subject variability pose significant challenges for MI-task recognition. This paper proposes a transfer learning assisted graph convolutional network (GCN) modeling approach for cross-dataset MI decoding, one of the most challenging issues in this field. In the experiments, a multi-channel dataset with 62 electrodes and a few-channel dataset with 8 electrodes are utilized for cross-dataset modeling. To harness multi-channel information, we utilize the GCN module to aggregate topological features. The pre-trained model is guided with few-channel signals as inputs through a knowledge distillation framework. Subsequently, the pre-trained model is adapted to the few-channel dataset using a transfer learning strategy with minimal data training. Experiment results show that the proposed model achieves 3.92% and 3.83% more accuracy improvement compared with state-of-the-art models in the cross-validation and cross-session scenario respectively, demonstrating the effectiveness of the proposed approach in cross-dataset MI-EEG decoding, thus enabling more effective MI-BCI applications. [ABSTRACT FROM AUTHOR]

Accurate traffic flow prediction is essential for intelligent transportation systems and urban planning. Traditional approaches that combine Transformer models with Graph Convolutional Networks (GCNs) or Convolutional Neural Networks (CNNs) often struggle to effectively integrate features with varying degrees of connectivity. As a result, graph-based problems do not fully utilise the capabilities of GCNs, while time-series problems fail to entirely leverage CNNs. To overcome these challenges, we introduce the Dual Heterogeneous Three-Stage Coupling and Multivariate Feature-Aware Learning Network (CHAMFormer). This architecture comprises three main components, each focusing on a distinct innovation: capturing fine-grained, short-range traffic patterns to manage immediate interactions and local bottlenecks; integrating mid-range spatial and temporal features to understand broader traffic interactions and ripple effects; and analysing complex, long-term traffic dynamics to anticipate and manage large-scale events and network behaviours across the entire system. These modules enhance GCN performance, enabling them to function more effectively alongside Transformers and Graph Neural Networks (GNNs). The CHAMFormer model incorporates a three-stage self-attention mechanism with a Skip-Connection method to improve the capture of detailed information without significantly increasing computational costs. By connecting low-level, intermediate-level, and high-level feature extractions, this model adapts well to changing traffic patterns, thereby enhancing multi-feature awareness and prediction accuracy. Extensive experiments using seven public datasets, both with and without predefined graph structures, and in multivariate and univariate scenarios demonstrate that CHAMFormer improves prediction accuracy by at least 10%–15%. To validate our proposed model, we also tested CHAMFormer in the energy domain, where it effectively handles time-series problems. Additionally, a sensitivity analysis confirms the model's predictability and interpretability, providing valuable insights for transportation policy and infrastructure development. [ABSTRACT FROM AUTHOR]

• Physics-integrated heterogeneous graph boosts feature richness and propagation. • The ME-GNN model enhances representation by leveraging distinct feature spaces. • The proposed feature fusion module improves the model's fitting capability. In practical engineering scenarios, the operating speed of mechanical equipment is intricate and variable. However, much of the existing research on intelligent fault diagnosis is conducted under constant speed conditions, with limited studies focusing on fault diagnosis in the presence of time-varying speeds. Moreover, the limitation of labeled data poses considerable obstacles for intelligent fault diagnosis methodologies. Therefore, a semi-supervised meta-path space extended graph neural network (ME-GNN) is proposed for fault diagnosis in the context of time-varying speeds and limited labeled samples. Firstly, a novel heterogeneous graph is proposed, which converts the nearest neighbor relationship between vibration data, fault information and variable speed information into a graph. This kind of graph not only integrates diverse physical information but also facilitates message passing and aggregation across heterogeneous data types. To obtain the feature information of heterogeneous graphs from different feature space, meta-path space extended graph convolution network is implemented to aggregate information from different attribute nodes. Finally, the designed feature fusion module effectively integrates node features and topological information, thereby further expanding the feature space and enhancing the diagnostic capability of the model. A series of comparative experiments validate that the proposed method surpasses existing fault diagnosis methods. [ABSTRACT FROM AUTHOR]

Dynamic spatio-temporal dependencies and temporal patterns in traffic series are critical factors affecting traffic forecasting accuracy. Due to the intrinsic challenges of incorporating explicit, logical knowledge into the implicit black-box learning process of neural networks, only a few methods effectively use prior knowledge to improve the learning of traffic forecasting. To tackle this problem, we introduce a new approach called Knowledge-augmented and Time-aware Graph Convolutional Network, namely KaTaGCN. At its core, we have created a knowledge-augmented module that boosts the diffusion weights between closely related adjacent nodes in graph learning. This is achieved by implementing a new loss function. Then, to learn the periodic implicit relationship between these timestamps and traffic signals, the weights and biases are chosen adaptively to be trained based on the timestamps of each sample. Finally, a gated spatio-temporal mapping module regresses high-dimensional embedded features from spatial and temporal dimensions. KaTaGCN is structured without any attention mechanisms or recurrent neural networks. Extensive experimental results on six real-world public traffic datasets demonstrate that KaTaGCN achieves an average improvement of 4.29% in forecasting performance compared with suboptimal results. • Propose a loss function that utilizes prior knowledge to guide adaptive graph learning. • Design a dynamic time-aware network to capture the time patterns on each node. • A concise and efficient framework without any attention mechanism or RNN structure. [ABSTRACT FROM AUTHOR]

Graph Convolutional Networks (GCNs) have attracted a lot of research interest in machine learning, and many variants have been proposed recently. In this paper, we take a step forward to establish a unified framework for convolution-based graph neural networks, aiming to provide a systematic view of different GCN variants and deep understanding of the relations among them. Our key idea is formulating the basic graph convolution operation as an optimization problem in the graph Fourier space. Under this framework, a variety of popular GCN models, including vanilla-GCNs, attention-based GCNs and topology-based GCNs, can be interpreted as a similar optimization problem but with different regularizers. This novel perspective enables a better understanding of the similarities and differences among many widely used GCNs, and may inspire new model designs. As a showcase, we present a novel regularization technique under the proposed framework to tackle the oversmoothing problem in graph convolution. The effectiveness of newly designed model is validated empirically. • A unified framework for GCNs, which interpret them as regularizers in Fourier space. • New insights on understanding GCNs and new directions to tackle common problems. • A once-for-all platform, where new designs can be pluged in with trivial adaptations. • A novel regularization technique to alleviate oversmoothing in graph learning. [ABSTRACT FROM AUTHOR]

The Gobi Desert has vast wilderness to utilize, and its renewable energy capacity experiencing rapid growth. To better allocate regulation resources for maintaining power balance and frequency regulation capacity, an islanded grid optimization model considering multi-timescale dispatch optimization is constructed for integrating chemical parks with thermal power units, energy storage, and green hydrogen production. In the day-ahead optimization stage, to improve the level of renewable energy consumption, the ladder-type regulation performance of thermal power units involved in deep peak load with frequency regulation capacity is derived and established. Moving to the real-time frequency regulation stage, a novel graph neural network-based technique with infeasible region modification is proposed to rapidly acquire the operational power scheme. The input features are the time series of total power command, regulation capacity, past operating power, and past mileage command. And the output features are the mileage commands received by the resources. Training data are generated through offline optimization with the genetic algorithm, which utilizes renewable energy generation and load data with a scale of three months in the Gobi Desert. In the one-month simulation test, the proposed method demonstrated a reduction in power deviation by approximately 32.7 % and an improvement in accuracy by roughly 16 %. [ABSTRACT FROM AUTHOR]

Graph convolutional neural networks (GCNs) with domain-specific feature aggregation capabilities have unique advantages in hyperspectral image (HSI) classification. However, current GCN-based approaches frequently encounter the issue of node characteristics being over-smoothed while aggregating in higher-order domains. Furthermore, GCN linear classifiers focus solely on sample separability and ignore the potential manifold information of graph features, resulting in a failure to fully investigate extracted features. To address these problems, we propose a global-local manifold embedding broad graph convolutional network (GLMBG) for HSI classification. In GLMBG, we designed two modules from feature extraction and classification perspectives: The graph convolutional edge feature fusion extractor (GEFF) and the broad classifier of global-local manifold embedding (BGLME). GEFF is designed to learn graph node and local edge features from HSI through GCN and recursive filtering, combining them in a weighted manner to construct fused graph features. BGLME is designed to replace traditional linear classifiers with broad learning classifiers through manifold regularized embedding, fully utilizing the global and local manifold discriminant information of graph node features. The combination of GEFF and BGLME effectively reduces over-smoothing of graph node features while maximizing the utilization of manifold discriminant information, hence improving model feature discriminative ability. Experimental evaluations of three commonly used hyperspectral datasets show that our method surpasses state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Traffic forecasting is an increasingly important research topic in the field of Intelligent Transportation Systems (ITS). In this field, prediction models based on Graph Convolution Networks (GCN) have become very popular. Most GCN-based models focus on constructing various optimized or dynamic road network graphs to represent the spatio-temporal correlation hidden in traffic data. However, these methods currently only consider the construction of a single improved road network graph and ignore the relationship of these existing optimized road network graphs. Therefore, in this paper, we propose a Contrastive Optimized Graph Convolution Network (COGCN) to connect two kinds of optimized road network graphs and maintain their global–local feature consistency through contrastive learning. The proposed COGCN model is evaluated in detail using four real traffic datasets: two traffic speed datasets and two traffic flow datasets. Experimental results show that COGCN improves forecasting accuracy by at least 2% on the two speed datasets and 9% on the two flow datasets compared to the existing state-of-the-art GCN-based methods. [ABSTRACT FROM AUTHOR]

Accurately predicting traffic flow characteristics is crucial for effective urban transportation management. Emergence of artificial intelligence has led to the surge of deep learning methods for short-term traffic forecast. Notably, Graph Convolutional Neural Networks (GCN) have demonstrated remarkable prediction accuracy by incorporating road network topology into deep neural networks. However, many existing GCN-based models are based on the premise that the graph network is static, which may fail to do justice in replicating the situations in the real World. On one hand, real road networks are dynamic and undergo changes such as road maintenance and traffic control, leading to altered network structures over time. On the other hand, relationships between road sections can fluctuate due to factors like traffic accidents, weather conditions, and other events, which can significantly impact traffic patterns and result in inaccurate predictions if a static network and static relationships between nodes are assumed. To address these challenges, we propose the spatiotemporal dual adaptive graph convolutional network (ST-DAGCN) model for spatiotemporal traffic prediction, which utilizes a dual-adaptive adjacency matrix comprising both a static and a dynamic graph structure learning matrix. The dual-adaptive mechanism can adaptively learn the global features and the local dynamic features of the traffic states by updating the correlations of nodes at each prediction step, while the gated recurrent unit (GRU), which is also a component of the model, extracts the temporal dependencies of traffic data. Through a comprehensive comparison analysis on two real-world traffic datasets, our model has achieved the highest prediction accuracy when compared to other advanced models. [ABSTRACT FROM AUTHOR]

• This study investigates information fusion methods for GNN-based traffic predictions, including their benefits and challenges. • A GNN-based information fusion technique improves traffic forecasting accuracy over conventional methods. • Integration of multi-source data improves traffic forecasting models. • Integration of GNNs with AI methods like evolutionary algorithms or reinforcement learning could be effective. • Hybrid models could improve overall performance and flexibility in challenging traffic situations. To improve forecasting accuracy and capture complex interactions within transportation networks, information fusion approaches are crucial for traffic predictions based on graph neural networks (GNNs). GNNs offer a potentially effective framework for capturing complex patterns and interactions among diverse elements, such as road segments and crossings, by considering both temporal and geographical dependencies. Although GNN-based traffic forecasting has recently been investigated in many studies, there is a need for comprehensive reviews that examine information fusion approaches for GNN-based traffic predictions, including an analysis of their benefits and challenges. This study addresses this knowledge gap and offers future insights into the potential advancements and developing fields of research in GNN-based fusion techniques, as well as their implications in urban planning and smart cities. Existing research demonstrates that the accuracy of traffic forecasting is substantially enhanced by information fusion techniques based on GNNs in comparison to more conventional approaches. By integrating information fusion methods with GNNs, the model is capable of capturing complex temporal and spatial relationships between various locations in a traffic network. Multi-source data integration benefits traffic forecasting models, including social events, weather conditions, real-time traffic sensor data, and historical traffic patterns. In addition, combining GNNs with other artificial intelligence (AI) methods like evolutionary algorithms or reinforcement learning could be an efficient strategy. With the potential to combine the best features of several methods, hybrid models could improve overall performance and flexibility in challenging traffic situations. [ABSTRACT FROM AUTHOR]

Cancer drug response (CDR) prediction is an important area of research that aims to personalize cancer therapy, optimizing treatment plans for maximum effectiveness while minimizing potential negative effects. Despite the advancements in Deep learning techniques, the effective integration of multi-omics data for drug response prediction remains challenging. In this paper, a regression method using Deep ResNet for CDR (DRN-CDR) prediction is proposed. We aim to explore the potential of considering sole cancer genes in drug response prediction. Here the multi-omics data such as gene expressions, mutation data, and methylation data along with the molecular structural information of drugs were integrated to predict the IC50 values of drugs. Drug features are extracted by employing a Uniform Graph Convolution Network, while Cell line features are extracted using a combination of Convolutional Neural Network and Fully Connected Networks. These features are then concatenated and fed into a deep ResNet for the prediction of IC50 values between Drug – Cell line pairs. The proposed method yielded higher Pearson's correlation coefficient (r p) of 0.7938 with lowest Root Mean Squared Error (RMSE) value of 0.92 when compared with similar methods of tCNNS, MOLI, DeepCDR, TGSA, NIHGCN, DeepTTA, GraTransDRP and TSGCNN. Further, when the model is extended to a classification problem to categorize drugs as sensitive or resistant, we achieved AUC and AUPR measures of 0.7623 and 0.7691, respectively. The drugs such as Tivozanib, SNX-2112, CGP-60474, PHA-665752, Foretinib etc., exhibited low median IC50 values and were found to be effective anti-cancer drugs. The case studies with different TCGA cancer types also revealed the effectiveness of SNX-2112, CGP-60474, Foretinib, Cisplatin, Vinblastine etc. This consistent pattern strongly suggests the effectiveness of the model in predicting CDR. [Display omitted] • Cancer Drug Response Prediction using multi-omics data of sole cancer genes with deep ResNet model. • Tivozanib, SNX-2112, CGP-60474, PHA-665752 and Foretinib with low IC50 values were found to be effective anti-cancer drugs. • Case studies with 24 TCGA cancer types revealed Cisplatin, Foretinib, CMK, Vinblastine as highly sensitive cancer drugs. [ABSTRACT FROM AUTHOR]

With the increase in the aging population of many countries, the prevalence of neovascular age-related macular degeneration (nAMD) is expected to increase. Morphological parameters such as intraretinal fluid (IRF), subretinal fluid (SRF), subretinal hyperreflective material (SHRM), and pigment epithelium detachment (PED) of spectral-domain optical coherence tomography (SD-OCT) images are vital markers for proper treatment of nAMD, especially to get the information of treatment response to determine the proper treatment interval and switching of anti-vascular endothelial growth factor (VEGF) agents. For the precise evaluation of the change in nAMD lesions and patient-specific treatment, quantitative evaluation of the lesions in the OCT volume scans is necessary. However, manual segmentation requires many resources, and the number of studies of automatic segmentation is increasing rapidly. Improving automated segmentation performance in SD-OCT visual results requires long-range contextual inference of spatial information between retinal lesions and layers. This paper proposes a GAGUNet (graph convolution network (GCN)-assisted attention-guided UNet) model with a novel global reasoning module considering these points. The dataset used in the main experiment of this study underwent rigorous review by a retinal specialist from Konkuk University Hospital in Korea, contributing to both data preprocessing and validation to ensure a qualitative assessment. We conducted experiments on the RETOUCH dataset as well to demonstrate the scalability of the proposed model. Overall, our model demonstrates superior performance over the baseline models in both quantitative and qualitative evaluations. • We demonstrate the limitations of the baseline in retinal segmentation. • We propose a novel model for automated retinal OCT segmentation. • The proposed model performs higher than the baseline in retinal segmentation. [ABSTRACT FROM AUTHOR]

Existing methods that use document-level information for event detection ignore the dependencies between sentences and also have shortcomings in modeling the dependencies among words. In this paper, we propose a novel H ierarchical G raph E nhanced E vent D etection (HGEED) framework to make full use of syntax and document information for the task of event detection. First, a sentence graph is used to model word-to-word dependencies, enriching the local information of words by incorporating syntactic features. Then, a document graph is built to model sentence-to-sentence dependencies, obtaining global semantic representations for word-level prediction. The experiment results on the widely used ACE 2005 and TAC KBP 2015 corpora show that our model can capture local and global information with dependencies and achieve significant improvements as compared to all baselines. [ABSTRACT FROM AUTHOR]

Deepfake videos are increasingly used in spreading fake news or propaganda having a serious impact on people and society. Traditional deepfake detectors exploit spatial and/or temporal inconsistencies to differentiate between real and fake videos. Owing to the rapidly advancing deepfake creation algorithms, the latest detectors have made use of physiological and biological facial features to create more generic solutions. Our proposed solution uses facial landmarks as the physiological identifiers of a person's face and through them develops a relationship between facial areas in normal speech and tampered speech. By creating a graph structure from the resulting sparse data, we were able to use a spatio-temporal graph convolutional network for classification, which has significantly fewer parameters and a shorter training time than traditional CNNs. We conducted a multitude of experiments on 3 datasets, utilizing spatio-temporal features. The results demonstrate that this technique has better generalization, and high performance compared to latest research in deepfake detection without the reliance on large deep learning models which are tuned to learning image discrepancies more than data patterns. Moreover, our use of facial landmark-based features with a graph structure paves the way for the development of an explainable AI model that can be relied on. • Deepfake video generation algorithms modify the data associated with facial landmarks to create fake images and videos which are used as a source for spreading fake news. • Facial landmarks represent an important structural representation of the face. • A feature vector can be found on each facial landmark to create a representation of the facial structure. • Facial structural data in form of landmarks and feature vectors can be classified using Graph CNNs to detect deepfakes since GCNs are better suited for sparse geometric data. [ABSTRACT FROM AUTHOR]

Dynamic facial expression recognition (DFER) is of great significance in promoting empathetic machines and metaverse technology. However, dynamic facial expression recognition (DFER) in the wild remains a challenging task, often constrained by complex lighting changes, frequent key-points occlusion, uncertain emotional peaks and severe imbalanced dataset categories. To tackle these problems, this paper presents a depth neural network model based on spatial key-points optimized region feature fusion and temporal self-attention. The method includes three parts: spatial feature extraction module, temporal feature extraction module and region feature fusion module. The intra-frame spatial feature extraction module is composed of the key-points graph convolution network (GCN) and a convolution network (CNN) branch to obtain the global and local feature vectors. The newly proposed region fusion strategy based on face spatial structure is used to obtain the spatial fusion feature of each frame. The inter-frame temporal feature extraction module uses multi-head self-attention model to obtain the temporal information of inter-frames. The experimental results show that our method achieves accuracy of 68.73%, 55.00%, 47.80%, and 47.44% on the DFEW, AFEW, FERV39k, and MAFW datasets. Ablation experiments showed that the GCN module, fusion module, and temporal module improved the accuracy on DFEW by 0.68%, 1.66%, and 3.25%, respectively. The method also achieves competitive results in terms of parameter quantity and inference speed, which demonstrates the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

• Deep learning mdoels explore pricise and efficient ash content estimation. • Scrutinizing color and texture features in neural network-based estimation. • Feature disentangled pipline and interpretable decision module were proposed. • CNN, ViT, hybrid model, and GCN were employed for feature representation. • Feature contribution analysis guides accuracy-driven model design. Determining coal ash content is paramount when evaluating coal quality and optimizing industrial processes. Conventional methods reliant on manual analysis prove exceedingly time-consuming and labor-intensive. The advent of deep learning has galvanized researchers to explore various models aimed at precision and efficiency in the coal industry. However, the selection of appropriate features assumes a pivotal role in achieving accurate estimation. This study meticulously scrutinizes the importance of color and texture features in estimating coal ash content using neural networks. The proposed framework for elucidating feature contributions encompasses the following steps: (1) A feature disentangled pipeline was employed to generate a color and texture dataset from the original dataset; (2) Harnessing convolutional neural network (CNN), vision Transformers(ViT), a hybrid model combing CNN and ViT, and graph convolution network (GCN) to learning feature representation for texture and color. 3) An interpretable decision module was employed to aggregate these two feature representations, thereby achieving an interpretable estimation of ash content. Experimental results and visualizations demonstrated the substantial importance of color in CNN, accounting for an impressive 64.77%, whereas the texture feature modestly contributed at 35.23%. The analysis of feature contributions assumes a crucial role in guiding the design of accuracy-driven models and in comprehending the inherent contributions or biases within deep learning architectures. [ABSTRACT FROM AUTHOR]

Multimodal fusion has the potential to significantly enhance model performance in the domain of Emotion Recognition in Conversations (ERC) by efficiently integrating information from diverse modalities. However, existing methods face challenges as they directly integrate information from different modalities, making it difficult to assess the individual impact of each modality during training and to capture nuanced fusion. To deal with it, we propose a novel framework named Fusing Pairwise Modalities for ERC. In this proposed method, the pairwise fusion technique is incorporated into multimodal fusion to enhance model performance, which enables each modality to contribute unique information, thereby facilitating a more comprehensive understanding of the emotional context. Additionally, a designed density loss is applied to characterise fused feature density, with a specific focus on mitigating redundancy in pairwise fusion methods. The density loss penalises feature density during training, contributing to a more efficient and effective fusion process. To validate the proposed framework, we conduct comprehensive experiments on two benchmark datasets, namely IEMOCAP and MELD. The results demonstrate the superior performance of our approach compared to state-of-the-art methods, indicating its effectiveness in addressing challenges related to multimodal fusion in the context of ERC. • Systematic pairwise management enhances multimodal fusion efficiency. • Designed density loss minimises feature redundancy, improving model robustness. • Demonstrates superior performance in multimodal fusion on IEMOCAP and MELD datasets. • Overcomes challenges associated with intricate inter-modal relationships in ERC. [ABSTRACT FROM AUTHOR]

The quality of traffic services depends on the accuracy and completeness of the collected traffic data. However,the existing traffic data imputation methods usually only rely on the predefined road network structure to capture the spatio-temporal features and only consider the imputation effect from a single perspective, which are very limited for imputation of different missing patterns of road traffic data. In this paper, we propose a novel deep learning framework called Hierarchical Spatio-temporal Graph Convolutional Neural Networks(HSTGCN) to impute traffic data,through the macro layer and the road layer. The model constructs macro graph of the road network based on the data temporal correlation clustering, which can mine the temporal dependencies of road traffic data from a hierarchical perspective. Besides, a temporal attention mechanism and adaptive adjacency matrix are introduced in the road layer to better extract the spatio-temporal information of the road traffic data. Finally, we use graph convolution neural networks to learn the spatio-temporal feature representations of the road layer and macro layer, which are then fused to achieve data imputation. To illustrate the efficient performance of the model, experiments are conducted on traffic data collected from California and Seattle. The proposed model performs better than the comparison model for traffic data imputation. • A full dynamic graph is built based on the urban road structure. • Extracting spatio-temporal features from multiple dimensions using hierarchical thinking. • Gate-GCN perform effective feature selection. [ABSTRACT FROM AUTHOR]

[Display omitted] • Employ contour features as geometric priors to preserve the outline details of lesions and explicitly model the correlation between region and contour. • Propose a parallel graph reasoning strategy to model the long-range relationship between adjacent pixels and analyze the global dependencies along the channel dimension of feature maps. • The improvements of the proposed method compare to the state-of-the-art models are validated on two public LiTS17 and 3DIRCADb datasets. The accurate detection and segmentation of liver cancers from abdominal CT scans is critical. However, segmenting liver tumors presents significant hurdles due to indistinct lesion boundaries and ignoring the correlation between target and outlines. In this paper, we propose the Parallel Graph Convolutional Network (PGC-Net), a completely novel segmentation framework for liver tumors. With regard to segmentation against constraints, we specifically use contour-induced parallel graph reasoning for quick yet efficient segmentation. First, we use a Pyramid Vision Transformer that has already been trained to extract multi-scale features of region and contour. In order to project the pixels into two distinct high-dimensional areas, we secondly use the parallel graph reasoning strategy, where the vertices are weighted in accordance with the geometric prior of the contour. Through the process of graph convolution, the complementary properties of region and contour also propagate the information. Finally, we project back to the original pixel space for the prediction using the refined features deduced from the graph. Experimental results on two available datasets, LiTS17 (with an average Dice score of 73.63%) and 3DIRCADb (with an average Dice score of 74.16%). Our framework focused on the interaction between two orthogonal graphs and contour information, which has the potential to improve the accuracy and efficiency of liver tumor segmentation. [ABSTRACT FROM AUTHOR]

Smart cities require advanced traffic management systems. Traffic forecasting is an essential task of the advanced transportation system. Traffic spatio-temporal data are often heterogeneous. Most existing traffic prediction models predominantly use separate components to extract the temporal and spatial features of traffic data. However, this overlooks the intrinsic connections between the spatio-temporal features of traffic data. To directly mine the spatio-temporal heterogeneity, this study constructs a global heterogeneous traffic spatio-temporal graph and proposes the Heterogeneous Traffic Spatio-Temporal Graph Convolution (HTSTGC). To reduce the complexity of the model, Simple Graph Convolution (SGC) is used to extract semi-structured meta-graph information. The receptive fields that capture temporal and spatial features can be flexibly adjusted separately through clever design, which can balance the performance and efficiency of the model. Finally, the feature fusion module applies Gated Graph Neural Network (GGNN) to fuse temporal and spatial features. The results on the PEMS datasets reveal that jointly modeling different types of relationships can improve the traffic prediction performance of the model. The proposed HTSTGC has better performance than the baseline methods in most cases. The research results can support urban traffic control, traffic pollution reduction, and sustainable urban development. • A spatio-temporal heterogeneous model was constructed for traffic prediction. • The heterogeneity of traffic flow can be captured from semi-structured information. • Temporal and spatial receptive fields can be flexibly adjusted respectively [ABSTRACT FROM AUTHOR]

Generally, Graph convolution network (GCN) utilizes the graph convolution operators and the softmax to extract the deep representation and make the prediction, respectively. Although GCN successfully represents the connectivity relationship among the nodes by aggregating the information on the graph, the softmax-based decision layer may result in suboptimal performance in semi-supervised learning with less label support due to ignoring the inner distribution of the graph nodes. Besides, the gradient descent will take thousands of interaction for optimization. To address the referred issues, we propose a novel graph deep model with a non-gradient decision layer for graph mining. Firstly, manifold learning is unified with label local-structure preservation to capture the topological information and make accurate predictions with limited label support. Moreover, it is theoretically proven to have analytical solutions and acts as a non-gradient decision layer in graph convolution networks. Particularly, a joint optimization method is designed for this graph model, which extremely accelerates the convergence of the model. Finally, extensive experiments show that the proposed model has achieved excellent performance compared to the current models. [Display omitted] • Unifying the orthogonal manifold with label local-structure preservation to mine the topological information of the deep embeddings and make more accurate predictions, the novel non-gradient graph decision layer is put forward. • With the assistance of the designed theorems, the non-gradient graph decision layer can be solved with an elegant analytical solution theoretically. • By embedding the analytical solution into the gradient descent, a joint optimization strategy is designed to jointly optimize the graph convolution network and the proposed non-gradient decision layer. [ABSTRACT FROM AUTHOR]

Traffic prediction is an important topic in intelligent transportation systems (ITSs) that can provide support for many traffic applications. However, accurate traffic prediction is a challenging task, and its difficulties mainly come from the complex spatial and temporal dependencies of traffic network data. Previous studies mainly focused on capturing the spatiotemporal dependencies of traffic data but ignored the temporal frequency features of traffic flows. In this paper, we design a novel multichannel spatial–temporal graph convolution network based on spectrum decomposition (MSDGCN) for traffic prediction. First, we decompose the original traffic flow series into low-frequency, mid-frequency and high-frequency components based on spectrum analysis. Then, based on ChebNet and LSTM, a prediction model with three channels is constructed to capture the low-frequency, mid-frequency and high-frequency features of the traffic flow. Finally, an attention layer is adopted to assign weights for different channels, enabling the model to focus on capturing the main features of the traffic flow. Compared to the state-of-the-art baselines, our model captures the distinct frequency components of the original traffic flow series in a more meticulous manner. Experimental results obtained on six real-world datasets indicate the excellent performance of our model. • We introduced DFT and spectrum analysis into traffic prediction. • An attention mechanism is utilized to reduce the adverse impact of local fluctuations. • A novel multichannel spatial–temporal graph convolution network is constructed. [ABSTRACT FROM AUTHOR]

Pose estimation can be used to analyse the behaviour of livestock, hence offering valuable insights into the physiological state of the livestock. Despite impressive progress in academic benchmarks, existing pose estimation methods do not meet the speed and accuracy trade-off requirements of industrial applications. Following the top-down paradigm, we propose GRMPose, a GCN-based real-time pose estimation framework for dairy goats. The framework adopts CSPNext as its backbone, which is highly efficient and potent for extracting features. In addition, a GCN-based coordination classification module (GCNCC) is proposed to improve the capability of extracting the structural pose information of the keypoint features. To assess the effectiveness of GRMPose, we built the DairyGoat dataset, which contains 2,108 images and 2,576 instances with different lighting conditions, different types of behaviours, and partially occluded dairy goat instance objects. Experimental results demonstrate that GRMPose achieves an AP of 87.48% on the DairyGoat dataset. Additionally, it exhibits a model inference latency of 13.82 ms, GFLOPs of 5.57, and parameters totalling 27.56 million. These results establish GRMPose as superior to other models like HRNet, Resnext, and MobileNetv2 in terms of speed-accuracy trade-off. • A GCN-based real-time pose estimation framework was proposed. • A GCN-based coordinate classification module can extract structural pose information. • This algorithm performs well in various scenarios, balancing speed and accuracy. [ABSTRACT FROM AUTHOR]

• We proposed a bi-channel convolution framework to estimate the CBT end-to-end; • We improved topological similarity constraints between the CBT and multi-view networks; • We constructed harmonic bases with the CBT to explore the alterations in anatomical features in the harmonic domain. Multi-view networks constructed from multiple brain functional or structural measures (e.g., fractional anisotropy, fiber number, and fiber length, etc.) describe brain connectivity from different views. The connectional brain template (CBT) of multi-view networks from a given population provides a standard graph template incorporating complementary information for analyzing intergroup differences. However, preserving the complex topology of multi-view networks and building the CBT end-to-end remains challenging. Hence, we proposed a bi-channel convolution framework, including a local connection channel (LCC) with edge-conditioned convolution layers and a global network channel (GNC) with graph convolution network, to estimate the CBT in an end-to-end manner. The LCC captured the local connection topology across individual networks, and the GNC learned the global network topology through the high-order population network built by the similarity between the node strength distributions of individual networks for each view. Additionally, we improved topological similarity constraint in the loss function by minimizing the KL divergence of the node strength and degree distributions between the CBT and multi-view networks. Compared with four existing methods, our CBT exhibited optimal centrality, topological structure retention ability and differentiation between the younger and older age groups. Using the harmonic bases of the CBT, we found that the total harmonic energy of cortical thickness presented significant differences among different age groups. Our method provided a standard common template for examining the structural connection changes in the spatial domain and the anatomical features alterations in the harmonic domain, thus distinguishing the different population groups. [ABSTRACT FROM AUTHOR]

In location-based social networks, calibrating a point-of-interest (POI) recommendation system is as important as its accuracy for improving user satisfaction. POI recommendation calibration is primarily classified as categorical or geographical calibration. Categorical calibration ensures that the recommended items are distributed proportionally among the past interest categories of the target user. When a target user checks 80 Chinese, 10 Japanese, and 10 French restaurants, a recommendation list with a ratio of 8:1:1 for Chinese, Japanese, and French restaurants can be reasonably expected. In addition to categorical calibration, geographical calibration has been proposed to increase user interest in the recommended results. Users have a high probability of revisiting locations in their subareas. Therefore, the POIs recommended in multiple subareas of interest are more suitable than those from one small and frequently visited subarea. However, improving the calibration and accuracy are conflicting tasks. To achieve high calibration while maintaining accuracy, previous studies proposed reranking-based techniques to rerank the candidate list and return POIs with high calibration. However, optimizing the calibration by reranking is independent of the basic-candidate-item generation model, resulting in a suboptimal system. To tackle the problem, we propose a novel sampling-based differentiation technique to merge the task of improving calibration into the GCN model training process and directly generate the final recommendation list. The model is flexible and can be applied to different domains, where a domain can be a subarea or category. In a three-layer GCN, the layer one represents the historical check-ins of the user, whereas layer three includes the candidate POIs from which the target user aggregates information. We trained the model to make the distribution of the POI domains at layer three approximated the distribution at layer one. Experimental results on Philadelphia and Tucson datasets confirmed that the proposed method outperforms all state-of-the-art GCN+ geo-reranking and GCN+ MCF baselines, improving Recall@ 5 from 0.0394 to 0.0412 (4.57%) and Jensen–Shannon measure (JS)@ 5 from 0.5931 to 0.6734 (13.54%) on the Philadelphia dataset and improving Recall@ 5 from 0.0495 to 0.0517 (4.40%) and JS@ 5 from 0.5869 to 0.6598 (12.42%) on the Tucson dataset for categorical calibration. The model was also tested in the geographical domain and a similar trend was observed. [ABSTRACT FROM AUTHOR]

Multimodal sentiment analysis aims to infer the sentiment of video bloggers from the features of multiple input modalities. However, there are problems such as signal noise and signal loss in the input phase and inefficient utilization of features in the modality fusion phase. To address these issues, this study proposes a feature-based restoration dynamic interaction network for multimodal sentiment analysis. Firstly, the idea of resampler and integration is employed to enhance visual and textual features during the input phase. Secondly, in the modal interaction phase, a dynamic routing network is employed. The network is centered on text modality and dynamically fuses visual and audio features. Finally, in the classification phase, multimodal representations are united to provide guidance for multimodal sentiment analysis. This study conducted experiments on the datasets MOSI, MOSEI and UR-FUNNY, which have 2199, 22856 and 16514 video segments respectively. The results show that the proposed method achieves an average improvement of about 1 point for three metrics on MOSI and 0.5 points for individual metrics on MOSEI compared to the state-of-the-art methods. Compared to other methods, the proposed approach achieve about 1 point improvement for individual metrics on UR-FUNNY dataset. • We introduce a dynamic interaction network for multimodal sentiment analysis. • We use a dynamic routing network to learn inter-modal consistency features. • We enhance the features of text and vision in the feature extraction phase. [ABSTRACT FROM AUTHOR]