Anomaly fluctuations in operating conditions, catalyst wear, crushing, and the deterioration of feedstock properties in fluid catalytic cracking (FCC) units can disrupt the normal circulating fluidization process of t...Anomaly fluctuations in operating conditions, catalyst wear, crushing, and the deterioration of feedstock properties in fluid catalytic cracking (FCC) units can disrupt the normal circulating fluidization process of the catalyst. Although several effective models have been proposed in previous research to address anomaly detection in chemical processes, most fail to adequately capture the spatial-temporal dependencies of multi-source, mixed-frequency information. In this study, an innovative multi-source mixed-frequency information fusion framework based on a spatial-temporal graph attention network (MIF-STGAT) is proposed to investigate the causes of FCC regenerator catalyst loss anomalies for guide onsite operational management, enhancing the long-term stability of FCC unit operations. First, a reconstruction-based dual-encoder-decoder framework is developed to facilitate the acquisition of mixed-frequency features and information fusion during the FCC regenerator catalyst loss process. Subsequently, a graph attention network and a multilayer long short-term memory network with a differential structure are integrated into the reconstruction-based dual-encoder-shared-decoder framework to capture the dynamic fluctuations and critical features associated with anomalies. Experimental results from the Chinese FCC industrial process demonstrate that MIF-STGAT achieves excellent accuracy and interpretability for anomaly detection.展开更多
It is known that long non-coding RNAs(lncRNAs)play vital roles in biological processes and contribute to the progression,development,and treatment of various diseases.Obviously,understanding associations between disea...It is known that long non-coding RNAs(lncRNAs)play vital roles in biological processes and contribute to the progression,development,and treatment of various diseases.Obviously,understanding associations between diseases and lncRNAs significantly enhances our ability to interpret disease mechanisms.Nevertheless,the process of determining lncRNA-disease associations is costly,labor-intensive,and time-consuming.Hence,it is expected to foster computational strategies to uncover lncRNA-disease relationships for further verification to save time and resources.In this study,a collaborative filtering and graph attention network-based LncRNA-Disease Association(CFGANLDA)method was nominated to expose potential lncRNA-disease associations.First,it takes into account the advantages of using biological information from multiple sources.Next,it uses a collaborative filtering technique in order to address the sparse data problem.It also employs a graph attention network to reinforce both linear and non-linear features of the associations to advance prediction performance.The computational results indicate that CFGANLDA gains better prediction performance compared to other state-of-the-art approaches.The CFGANLDA’s area under the receiver operating characteristic curve(AUC)metric is 0.9835,whereas its area under the precision-recall curve(AUPR)metric is 0.9822.Statistical analysis using 10-fold cross-validation experiments proves that these metrics are significant.Furthermore,three case studies on prostate,liver,and stomach cancers attest to the validity of CFGANLDA performance.As a result,CFGANLDA method proves to be a valued tool for lncRNA-disease association prediction.展开更多
Pedestrian trajectory prediction is pivotal and challenging in applications such as autonomous driving,social robotics,and intelligent surveillance systems.Pedestrian trajectory is governed not only by individual inte...Pedestrian trajectory prediction is pivotal and challenging in applications such as autonomous driving,social robotics,and intelligent surveillance systems.Pedestrian trajectory is governed not only by individual intent but also by interactions with surrounding agents.These interactions are critical to trajectory prediction accuracy.While prior studies have employed Convolutional Neural Networks(CNNs)and Graph Convolutional Networks(GCNs)to model such interactions,these methods fail to distinguish varying influence levels among neighboring pedestrians.To address this,we propose a novel model based on a bidirectional graph attention network and spatio-temporal graphs to capture dynamic interactions.Specifically,we construct temporal and spatial graphs encoding the sequential evolution and spatial proximity among pedestrians.These features are then fused and processed by the Bidirectional Graph Attention Network(Bi-GAT),which models the bidirectional interactions between the target pedestrian and its neighbors.The model computes node attention weights(i.e.,similarity scores)to differentially aggregate neighbor information,enabling fine-grained interaction representations.Extensive experiments conducted on two widely used pedestrian trajectory prediction benchmark datasets demonstrate that our approach outperforms existing state-of-theartmethods regarding Average Displacement Error(ADE)and Final Displacement Error(FDE),highlighting its strong prediction accuracy and generalization capability.展开更多
Counterfeit agricultural products pose a significant challenge to global food security and economic stability, necessitating advanced detection mechanisms to ensure authenticity and quality. To address this pressing i...Counterfeit agricultural products pose a significant challenge to global food security and economic stability, necessitating advanced detection mechanisms to ensure authenticity and quality. To address this pressing issue, we introduce iGFruit, an innovative model designed to enhance the detection of counterfeit agricultural products by integrating multimodal data processing. Our approach utilizes both image and text data for comprehensive feature extraction, employing advanced backbone models such as Vision Transformer (ViT), Normalizer-Free Network (NFNet), and Bidirectional Encoder Representations from Transformers (BERT). These extracted features are fused and processed using a Graph Attention Network (GAT) to capture intricate relationships within the multimodal data. The resulting fused representation is subsequently classified to detect counterfeit products with high precision. We validate the effectiveness of iGFruit through extensive experiments on two datasets: the publicly available MIT-States dataset and the proprietary TLU-States dataset, achieving state-of-the-art performance on both benchmarks. Specifically, iGFruit demonstrates an improvement of over 3% in average accuracy compared to baseline models, all while maintaining computational efficiency during inference. This work underscores the necessity and innovativeness of integrating graph-based feature learning to tackle the critical issue of counterfeit agricultural product detection.展开更多
Continuous sign language recognition(CSLR)is challenging due to the complexity of video background,hand gesture variability,and temporal modeling difficulties.This work proposes a CSLR method based on a spatialtempora...Continuous sign language recognition(CSLR)is challenging due to the complexity of video background,hand gesture variability,and temporal modeling difficulties.This work proposes a CSLR method based on a spatialtemporal graph attention network to focus on essential features of video series.The method considers local details of sign language movements by taking the information on joints and bones as inputs and constructing a spatialtemporal graph to reflect inter-frame relevance and physical connections between nodes.The graph-based multihead attention mechanism is utilized with adjacent matrix calculation for better local-feature exploration,and short-term motion correlation modeling is completed via a temporal convolutional network.We adopted BLSTM to learn the long-termdependence and connectionist temporal classification to align the word-level sequences.The proposed method achieves competitive results regarding word error rates(1.59%)on the Chinese Sign Language dataset and the mean Jaccard Index(65.78%)on the ChaLearn LAP Continuous Gesture Dataset.展开更多
Recent advances in spatially resolved transcriptomic technologies have enabled unprecedented opportunities to elucidate tissue architecture and function in situ.Spatial transcriptomics can provide multimodal and compl...Recent advances in spatially resolved transcriptomic technologies have enabled unprecedented opportunities to elucidate tissue architecture and function in situ.Spatial transcriptomics can provide multimodal and complementary information simultaneously,including gene expression profiles,spatial locations,and histology images.However,most existing methods have limitations in efficiently utilizing spatial information and matched high-resolution histology images.To fully leverage the multi-modal information,we propose a SPAtially embedded Deep Attentional graph Clustering(SpaDAC)method to identify spatial domains while reconstructing denoised gene expression profiles.This method can efficiently learn the low-dimensional embeddings for spatial transcriptomics data by constructing multi-view graph modules to capture both spatial location connectives and morphological connectives.Benchmark results demonstrate that SpaDAC outperforms other algorithms on several recent spatial transcriptomics datasets.SpaDAC is a valuable tool for spatial domain detection,facilitating the comprehension of tissue architecture and cellular microenvironment.The source code of SpaDAC is freely available at Github(https://github.com/huoyuying/SpaDAC.git).展开更多
PM2.5 concentration prediction is of great significance to environmental protection and human health.Achieving accurate prediction of PM2.5 concentration has become an important research task.However,PM2.5 pollutants ...PM2.5 concentration prediction is of great significance to environmental protection and human health.Achieving accurate prediction of PM2.5 concentration has become an important research task.However,PM2.5 pollutants can spread in the earth’s atmosphere,causing mutual influence between different cities.To effectively capture the air pollution relationship between cities,this paper proposes a novel spatiotemporal model combining graph attention neural network(GAT)and gated recurrent unit(GRU),named GAT-GRU for PM2.5 concentration prediction.Specifically,GAT is used to learn the spatial dependence of PM2.5 concentration data in different cities,and GRU is to extract the temporal dependence of the long-term data series.The proposed model integrates the learned spatio-temporal dependencies to capture long-term complex spatio-temporal features.Considering that air pollution is related to the meteorological conditions of the city,the knowledge acquired from meteorological data is used in the model to enhance PM2.5 prediction performance.The input of the GAT-GRU model consists of PM2.5 concentration data and meteorological data.In order to verify the effectiveness of the proposed GAT-GRU prediction model,this paper designs experiments on real-world datasets compared with other baselines.Experimental results prove that our model achieves excellent performance in PM2.5 concentration prediction.展开更多
Numerous works prove that existing neighbor-averaging graph neural networks(GNNs)cannot efficiently catch structure features,and many works show that injecting structure,distance,position,or spatial features can signi...Numerous works prove that existing neighbor-averaging graph neural networks(GNNs)cannot efficiently catch structure features,and many works show that injecting structure,distance,position,or spatial features can significantly improve the performance of GNNs,however,injecting high-level structure and distance into GNNs is an intuitive but untouched idea.This work sheds light on this issue and proposes a scheme to enhance graph attention networks(GATs)by encoding distance and hop-wise structure statistics.Firstly,the hop-wise structure and distributional distance information are extracted based on several hop-wise ego-nets of every target node.Secondly,the derived structure information,distance information,and intrinsic features are encoded into the same vector space and then added together to get initial embedding vectors.Thirdly,the derived embedding vectors are fed into GATs,such as GAT and adaptive graph diffusion network(AGDN)to get the soft labels.Fourthly,the soft labels are fed into correct and smooth(C&S)to conduct label propagation and get final predictions.Experiments show that the distance and hop-wise structures encoding enhanced graph attention networks(DHSEGATs)achieve a competitive result.展开更多
Considering the nonlinear structure and spatial-temporal correlation of traffic network,and the influence of potential correlation between nodes of traffic network on the spatial features,this paper proposes a traffic...Considering the nonlinear structure and spatial-temporal correlation of traffic network,and the influence of potential correlation between nodes of traffic network on the spatial features,this paper proposes a traffic speed prediction model based on the combination of graph attention network with self-adaptive adjacency matrix(SAdpGAT)and bidirectional gated recurrent unit(BiGRU).First-ly,the model introduces graph attention network(GAT)to extract the spatial features of real road network and potential road network respectively in spatial dimension.Secondly,the spatial features are input into BiGRU to extract the time series features.Finally,the prediction results of the real road network and the potential road network are connected to generate the final prediction results of the model.The experimental results show that the prediction accuracy of the proposed model is im-proved obviously on METR-LA and PEMS-BAY datasets,which proves the advantages of the pro-posed spatial-temporal model in traffic speed prediction.展开更多
The growing prevalence of knowledge reasoning using knowledge graphs(KGs)has substantially improved the accuracy and efficiency of intelligent medical diagnosis.However,current models primarily integrate electronic me...The growing prevalence of knowledge reasoning using knowledge graphs(KGs)has substantially improved the accuracy and efficiency of intelligent medical diagnosis.However,current models primarily integrate electronic medical records(EMRs)and KGs into the knowledge reasoning process,ignoring the differing significance of various types of knowledge in EMRs and the diverse data types present in the text.To better integrate EMR text information,we propose a novel intelligent diagnostic model named the Graph ATtention network incorporating Text representation in knowledge reasoning(GATiT),which comprises text representation,subgraph construction,knowledge reasoning,and diagnostic classification.In the text representation process,GATiT uses a pre-trained model to obtain text representations of the EMRs and additionally enhances embeddings by including chief complaint information and numerical information in the input.In the subgraph construction process,GATiT constructs text subgraphs and disease subgraphs from the KG,utilizing EMR text and the disease to be diagnosed.To differentiate the varying importance of nodes within the subgraphs features such as node categories,relevance scores,and other relevant factors are introduced into the text subgraph.Themessage-passing strategy and attention weight calculation of the graph attention network are adjusted to learn these features in the knowledge reasoning process.Finally,in the diagnostic classification process,the interactive attention-based fusion method integrates the results of knowledge reasoning with text representations to produce the final diagnosis results.Experimental results on multi-label and single-label EMR datasets demonstrate the model’s superiority over several state-of-theart methods.展开更多
Automatic text summarization(ATS)plays a significant role in Natural Language Processing(NLP).Abstractive summarization produces summaries by identifying and compressing the most important information in a document.Ho...Automatic text summarization(ATS)plays a significant role in Natural Language Processing(NLP).Abstractive summarization produces summaries by identifying and compressing the most important information in a document.However,there are only relatively several comprehensively evaluated abstractive summarization models that work well for specific types of reports due to their unstructured and oral language text characteristics.In particular,Chinese complaint reports,generated by urban complainers and collected by government employees,describe existing resident problems in daily life.Meanwhile,the reflected problems are required to respond speedily.Therefore,automatic summarization tasks for these reports have been developed.However,similar to traditional summarization models,the generated summaries still exist problems of informativeness and conciseness.To address these issues and generate suitably informative and less redundant summaries,a topic-based abstractive summarization method is proposed to obtain global and local features.Additionally,a heterogeneous graph of the original document is constructed using word-level and topic-level features.Experiments and analyses on public review datasets(Yelp and Amazon)and our constructed dataset(Chinese complaint reports)show that the proposed framework effectively improves the performance of the abstractive summarization model for Chinese complaint reports.展开更多
Advanced carbon emission factors of a power grid can provide users with effective carbon reduction advice,which is of immense importance in mobilizing the entire society to reduce carbon emissions.The method of calcul...Advanced carbon emission factors of a power grid can provide users with effective carbon reduction advice,which is of immense importance in mobilizing the entire society to reduce carbon emissions.The method of calculating node carbon emission factors based on the carbon emissions flow theory requires real-time parameters of a power grid.Therefore,it cannot provide carbon factor information beforehand.To address this issue,a prediction model based on the graph attention network is proposed.The model uses a graph structure that is suitable for the topology of the power grid and designs a supervised network using the loads of the grid nodes and the corresponding carbon factor data.The network extracts features and transmits information more suitable for the power system and can flexibly adjust the equivalent topology,thereby increasing the diversity of the structure.Its input and output data are simple,without the power grid parameters.We demonstrated its effect by testing IEEE-39 bus and IEEE-118 bus systems with average error rates of 2.46%and 2.51%.展开更多
The high proportion of renewable energy integration and the dynamic changes in grid topology necessitate the enhancement of voltage/var control(VVC)to manage voltage fluctuations more rapidly.Traditional model-based c...The high proportion of renewable energy integration and the dynamic changes in grid topology necessitate the enhancement of voltage/var control(VVC)to manage voltage fluctuations more rapidly.Traditional model-based control algorithms are becoming increasingly incompetent for VVC due to their high model dependence and slow online computation speed.To alleviate these issues,this paper introduces a graph attention network(GAT)based deep reinforcement learning for VVC of topologically variable power system.Firstly,combining the physical information of the actual power grid,a physics-informed GAT is proposed and embedded into the proximal policy optimization(PPO)algorithm.The GAT-PPO algorithm can capture topological and spatial correlations among the node features to tackle topology changes.To address the slow training,the Relief F-S algorithm identifies critical state variables,significantly reducing the dimensionality of state space.Then,the training samples retained in the experience buffer are designed to mitigate the sparse reward issue.Finally,the validation on the modified IEEE 39-bus system and an actual power grid demonstrates superior performance of the proposed algorithm compared with state-of-the-art algorithms,including PPO algorithm and twin delayed deep deterministic policy gradient(TD3)algorithm.The proposed algorithm exhibits enhanced convergence during training,faster solution speed,and improved VVC performance,even in scenarios involving grid topology changes and increased renewable energy integration.Meanwhile,in the adopted cases,the network loss is reduced by 6.9%,10.8%,and 7.7%,respectively,demonstrating favorable economic outcomes.展开更多
Imaging mass cytometry(IMC)enables the high-resolution spatial profiling of tumor microenvironment,but its clinical utility for prospective prediction remains underdeveloped.In this study,we integrated IMC into a clin...Imaging mass cytometry(IMC)enables the high-resolution spatial profiling of tumor microenvironment,but its clinical utility for prospective prediction remains underdeveloped.In this study,we integrated IMC into a clinical trial of hepatocellular carcinoma(HCC)patients undergoing combination therapy with programmed death-1 blockade and transarterial chemoembolization.We analyzed 281 regions of interest from 43 patients using a custom 40-marker IMC panel and developed a novel superpixel-based graph attention network,IMCSGAT,to model spatial cell interactions within the tumor microenvironment.IMCSGAT enabled accurate multitask prediction of key clinical features,including Barcelona Clinic Liver Cancer stage,trabecular histologic subtype,and treatment response.Compared to state-of-the-art methods,IMCSGAT achieved superior performance across all classification tasks.Spatial interaction analysis revealed that resident macrophage-centered interactions,particularly those with NK and T cells,were enriched in responders and predictive of therapeutic outcome.These findings were validated in a murine HCC model,reinforcing the role of innate immune architecture in shaping the treatment response.This study establishes IMCSGAT as a powerful spatial learning framework for high-dimensional IMC data,with potential applications in clinical outcome prediction and personalized therapy design for HCC.Our results provide a blueprint for the broader use of spatial analytics in precision oncology.展开更多
For permanent faults(PF)in the power communication network(PCN),such as link interruptions,the timesensitive networking(TSN)relied on by PCN,typically employs spatial redundancy fault-tolerance methods to keep service...For permanent faults(PF)in the power communication network(PCN),such as link interruptions,the timesensitive networking(TSN)relied on by PCN,typically employs spatial redundancy fault-tolerance methods to keep service stability and reliability,which often limits TSN scheduling performance in fault-free ideal states.So this paper proposes a graph attention residual network-based routing and fault-tolerant scheduling mechanism(GRFS)for data flow in PCN,which specifically includes a communication system architecture for integrated terminals based on a cyclic queuing and forwarding(CQF)model and fault recovery method,which reduces the impact of faults by simplified scheduling configurations of CQF and fault-tolerance of prioritizing the rerouting of faulty time-sensitive(TS)flows;considering that PF leading to changes in network topology is more appropriately solved by doing routing and time slot injection decisions hop-by-hop,and that reasonable network load can reduce the damage caused by PF and reserve resources for the rerouting of faulty TS flows,an optimization model for joint routing and scheduling is constructed with scheduling success rate as the objective,and with traffic latency and network load as constraints;to catch changes in TSN topology and traffic load,a D3QN algorithm based on a multi-head graph attention residual network(MGAR)is designed to solve the problem model,where the MGAR based encoder reconstructs the TSN status into feature embedding vectors,and a dueling network decoder performs decoding tasks on the reconstructed feature embedding vectors.Simulation results show that GRFS outperforms heuristic fault-tolerance algorithms and other benchmark schemes by approximately 10%in routing and scheduling success rate in ideal states and 5%in rerouting and rescheduling success rate in fault states.展开更多
Dialogue State Tracking(DST)is a critical component of task-oriented spoken dialogue systems(SDS),tasked with maintaining an accurate representation of the conversational state by predicting slots and their correspond...Dialogue State Tracking(DST)is a critical component of task-oriented spoken dialogue systems(SDS),tasked with maintaining an accurate representation of the conversational state by predicting slots and their corresponding values.Recent advances leverage Large Language Models(LLMs)with prompt-based tuning to improve tracking accuracy and efficiency.However,these approaches often incur substantial computational and memory overheads and typically address slot extraction implicitly within prompts,without explicitly modeling the complex dependencies between slots and values.In this work,we propose PUGG,a novel DST framework that constructs schema-driven prompts to fine-tune GPT-2 and utilizes its tokenizer to implement a memory encoder.PUGG explicitly extracts slot values via GPT-2 and employs Graph Attention Networks(GATs)to model and reason over the intricate relationships between slots and their associated values.We evaluate PUGG on four publicly available datasets,where it achieves stateof-the-art performance across multiple evaluation metrics,highlighting its robustness and generalizability in diverse conversational scenarios.Our results indicate that the integration of GPT-2 substantially reduces model complexity and memory consumption by streamlining key processes.Moreover,prompt tuning enhances the model’s flexibility and precision in extracting relevant slot-value pairs,while the incorporation of GATs facilitates effective relational reasoning,leading to improved dialogue state representations.展开更多
Most existing knowledge graph relationship prediction methods are unable to capture the complex information of multi-relational knowledge graphs,thus overlooking key details contained in different entity pairs and mak...Most existing knowledge graph relationship prediction methods are unable to capture the complex information of multi-relational knowledge graphs,thus overlooking key details contained in different entity pairs and making it difficult to aggregate more complex relational features.Moreover,the insufficient capture of multi-hop relational information limits the processing capability of the global structure of the graph and reduces the accuracy of the knowledge graph completion task.This paper uses graph neural networks to construct new message functions for different relations,which can be defined as the rotation from the source entity to the target entity in the complex vector space for each relation,thereby improving the relation perception.To further enrich the relational diversity of different entities,we capture themulti-hop structural information in complex graph structure relations by incorporating two-hop relations for each entity and adding auxiliary edges to various relation combinations in the knowledge graph,thereby aggregating more complex relations and improving the reasoning ability of complex relational information.To verify the effectiveness of the proposed method,we conducted experiments on the WN18RR and FB15k-237 standard datasets.The results show that the method proposed in this study outperforms most existing methods.展开更多
Quantum error-correcting codes are essential for fault-tolerant quantum computing,as they effectively detect and correct noise-induced errors by distributing information across multiple physical qubits.The subsystem s...Quantum error-correcting codes are essential for fault-tolerant quantum computing,as they effectively detect and correct noise-induced errors by distributing information across multiple physical qubits.The subsystem surface code with three-qubit check operators demonstrates significant application potential due to its simplified measurement operations and low logical error rates.However,the existing minimum-weight perfect matching(MWPM)algorithm exhibits high computational complexity and lacks flexibility in large-scale systems.Therefore,this paper proposes a decoder based on a graph attention network(GAT),representing error syndromes as undirected graphs with edge weights,and employing a multihead attention mechanism to efficiently aggregate node features and enable parallel computation.Compared to MWPM,the GAT decoder exhibits linear growth in computational complexity,adapts to different quantum code structures,and demonstrates stronger robustness under high physical error rates.The experimental results demonstrate that the proposed decoder achieves an overall accuracy of 89.95%under various small code lattice sizes(L=2,3,4,5),with the logical error rate threshold increasing to 0.0078,representing an improvement of approximately 13.04%compared to the MWPM decoder.This result significantly outperforms traditional methods,showcasing superior performance under small code lattice sizes and providing a more efficient decoding solution for large-scale quantum error correction.展开更多
Due to coexistence of huge number of structural isomers,global search for the ground-state structures of atomic clusters is a challenging issue.The difficulty also originates from the computational cost of ab initio m...Due to coexistence of huge number of structural isomers,global search for the ground-state structures of atomic clusters is a challenging issue.The difficulty also originates from the computational cost of ab initio methods for describing the potential energy surface.Recently,machine learning techniques have been widely utilized to accelerate materials discovery and molecular simulation.Compared to the commonly used artificial neural network,graph network is naturally suitable for clusters with flexible geometric environment of each atom.Herein we develop a cluster graph attention network(CGANet)by aggregating information of neighboring vertices and edges using attention mechanism,which can precisely predict the binding energy and force of silver clusters with root mean square error of 5.4 meV/atom and mean absolute error of 42.3 meV/Å,respectively.As a proof-of-concept,we have performed global optimization of mediumsized Agn clusters(n=14–26)by combining CGANet and genetic algorithm.The reported ground-state structures for n=14–21,have been successfully reproduced,while entirely new lowest-energy structures are obtained for n=22–26.In addition to the description of potential energy surface,the CGANet is also applied to predict the electronic properties of clusters,such as HOMO energy and HOMO-LUMO gap.With accuracy comparable to ab initio methods and acceleration by at least two orders of magnitude,CGANet holds great promise in global search of lowest-energy structures of large clusters and inverse design of functional clusters.展开更多
Event detection(ED)seeks to recognize event triggers and classify them into the predefined event types.Chinese ED is formulated as a character-level task owing to the uncertain word boundaries.Prior methods try to inc...Event detection(ED)seeks to recognize event triggers and classify them into the predefined event types.Chinese ED is formulated as a character-level task owing to the uncertain word boundaries.Prior methods try to incorpo-rate word-level information into characters to enhance their semantics.However,they experience two problems.First,they fail to incorporate word-level information into each character the word encompasses,causing the insufficient word-charac-ter interaction problem.Second,they struggle to distinguish events of similar types with limited annotated instances,which is called the event confusing problem.This paper proposes a novel model named Label-Aware Heterogeneous Graph Attention Network(L-HGAT)to address these two problems.Specifically,we first build a heterogeneous graph of two node types and three edge types to maximally preserve word-character interactions,and then deploy a heterogeneous graph attention network to enhance the semantic propagation between characters and words.Furthermore,we design a pushing-away game to enlarge the predicting gap between the ground-truth event type and its confusing counterpart for each character.Experimental results show that our L-HGAT model consistently achieves superior performance over prior competitive methods.展开更多
基金supported by the Innovative Research Group Project of the National Natural Science Foundation of China(22021004)Sinopec Major Science and Technology Projects(321123-1).
文摘Anomaly fluctuations in operating conditions, catalyst wear, crushing, and the deterioration of feedstock properties in fluid catalytic cracking (FCC) units can disrupt the normal circulating fluidization process of the catalyst. Although several effective models have been proposed in previous research to address anomaly detection in chemical processes, most fail to adequately capture the spatial-temporal dependencies of multi-source, mixed-frequency information. In this study, an innovative multi-source mixed-frequency information fusion framework based on a spatial-temporal graph attention network (MIF-STGAT) is proposed to investigate the causes of FCC regenerator catalyst loss anomalies for guide onsite operational management, enhancing the long-term stability of FCC unit operations. First, a reconstruction-based dual-encoder-decoder framework is developed to facilitate the acquisition of mixed-frequency features and information fusion during the FCC regenerator catalyst loss process. Subsequently, a graph attention network and a multilayer long short-term memory network with a differential structure are integrated into the reconstruction-based dual-encoder-shared-decoder framework to capture the dynamic fluctuations and critical features associated with anomalies. Experimental results from the Chinese FCC industrial process demonstrate that MIF-STGAT achieves excellent accuracy and interpretability for anomaly detection.
基金supported by the Vietnam Ministry of Education and Training under project code B2023-SPH-14。
文摘It is known that long non-coding RNAs(lncRNAs)play vital roles in biological processes and contribute to the progression,development,and treatment of various diseases.Obviously,understanding associations between diseases and lncRNAs significantly enhances our ability to interpret disease mechanisms.Nevertheless,the process of determining lncRNA-disease associations is costly,labor-intensive,and time-consuming.Hence,it is expected to foster computational strategies to uncover lncRNA-disease relationships for further verification to save time and resources.In this study,a collaborative filtering and graph attention network-based LncRNA-Disease Association(CFGANLDA)method was nominated to expose potential lncRNA-disease associations.First,it takes into account the advantages of using biological information from multiple sources.Next,it uses a collaborative filtering technique in order to address the sparse data problem.It also employs a graph attention network to reinforce both linear and non-linear features of the associations to advance prediction performance.The computational results indicate that CFGANLDA gains better prediction performance compared to other state-of-the-art approaches.The CFGANLDA’s area under the receiver operating characteristic curve(AUC)metric is 0.9835,whereas its area under the precision-recall curve(AUPR)metric is 0.9822.Statistical analysis using 10-fold cross-validation experiments proves that these metrics are significant.Furthermore,three case studies on prostate,liver,and stomach cancers attest to the validity of CFGANLDA performance.As a result,CFGANLDA method proves to be a valued tool for lncRNA-disease association prediction.
基金funded by the National Natural Science Foundation of China,grant number 624010funded by the Natural Science Foundation of Anhui Province,grant number 2408085QF202+1 种基金funded by the Anhui Future Technology Research Institute Industry Guidance Fund Project,grant number 2023cyyd04funded by the Project of Research of Anhui Polytechnic University,grant number Xjky2022150.
文摘Pedestrian trajectory prediction is pivotal and challenging in applications such as autonomous driving,social robotics,and intelligent surveillance systems.Pedestrian trajectory is governed not only by individual intent but also by interactions with surrounding agents.These interactions are critical to trajectory prediction accuracy.While prior studies have employed Convolutional Neural Networks(CNNs)and Graph Convolutional Networks(GCNs)to model such interactions,these methods fail to distinguish varying influence levels among neighboring pedestrians.To address this,we propose a novel model based on a bidirectional graph attention network and spatio-temporal graphs to capture dynamic interactions.Specifically,we construct temporal and spatial graphs encoding the sequential evolution and spatial proximity among pedestrians.These features are then fused and processed by the Bidirectional Graph Attention Network(Bi-GAT),which models the bidirectional interactions between the target pedestrian and its neighbors.The model computes node attention weights(i.e.,similarity scores)to differentially aggregate neighbor information,enabling fine-grained interaction representations.Extensive experiments conducted on two widely used pedestrian trajectory prediction benchmark datasets demonstrate that our approach outperforms existing state-of-theartmethods regarding Average Displacement Error(ADE)and Final Displacement Error(FDE),highlighting its strong prediction accuracy and generalization capability.
文摘Counterfeit agricultural products pose a significant challenge to global food security and economic stability, necessitating advanced detection mechanisms to ensure authenticity and quality. To address this pressing issue, we introduce iGFruit, an innovative model designed to enhance the detection of counterfeit agricultural products by integrating multimodal data processing. Our approach utilizes both image and text data for comprehensive feature extraction, employing advanced backbone models such as Vision Transformer (ViT), Normalizer-Free Network (NFNet), and Bidirectional Encoder Representations from Transformers (BERT). These extracted features are fused and processed using a Graph Attention Network (GAT) to capture intricate relationships within the multimodal data. The resulting fused representation is subsequently classified to detect counterfeit products with high precision. We validate the effectiveness of iGFruit through extensive experiments on two datasets: the publicly available MIT-States dataset and the proprietary TLU-States dataset, achieving state-of-the-art performance on both benchmarks. Specifically, iGFruit demonstrates an improvement of over 3% in average accuracy compared to baseline models, all while maintaining computational efficiency during inference. This work underscores the necessity and innovativeness of integrating graph-based feature learning to tackle the critical issue of counterfeit agricultural product detection.
基金supported by the Key Research&Development Plan Project of Shandong Province,China(No.2017GGX10127).
文摘Continuous sign language recognition(CSLR)is challenging due to the complexity of video background,hand gesture variability,and temporal modeling difficulties.This work proposes a CSLR method based on a spatialtemporal graph attention network to focus on essential features of video series.The method considers local details of sign language movements by taking the information on joints and bones as inputs and constructing a spatialtemporal graph to reflect inter-frame relevance and physical connections between nodes.The graph-based multihead attention mechanism is utilized with adjacent matrix calculation for better local-feature exploration,and short-term motion correlation modeling is completed via a temporal convolutional network.We adopted BLSTM to learn the long-termdependence and connectionist temporal classification to align the word-level sequences.The proposed method achieves competitive results regarding word error rates(1.59%)on the Chinese Sign Language dataset and the mean Jaccard Index(65.78%)on the ChaLearn LAP Continuous Gesture Dataset.
基金supported by National Natural Science Foundation of China(62003028).X.L.was supported by a Scholarship from the China Scholarship Council.
文摘Recent advances in spatially resolved transcriptomic technologies have enabled unprecedented opportunities to elucidate tissue architecture and function in situ.Spatial transcriptomics can provide multimodal and complementary information simultaneously,including gene expression profiles,spatial locations,and histology images.However,most existing methods have limitations in efficiently utilizing spatial information and matched high-resolution histology images.To fully leverage the multi-modal information,we propose a SPAtially embedded Deep Attentional graph Clustering(SpaDAC)method to identify spatial domains while reconstructing denoised gene expression profiles.This method can efficiently learn the low-dimensional embeddings for spatial transcriptomics data by constructing multi-view graph modules to capture both spatial location connectives and morphological connectives.Benchmark results demonstrate that SpaDAC outperforms other algorithms on several recent spatial transcriptomics datasets.SpaDAC is a valuable tool for spatial domain detection,facilitating the comprehension of tissue architecture and cellular microenvironment.The source code of SpaDAC is freely available at Github(https://github.com/huoyuying/SpaDAC.git).
基金Authors The research project is partially supported by National Natural ScienceFoundation of China under Grant No. 62072015, U19B2039, U1811463National Key R&D Programof China 2018YFB1600903.
文摘PM2.5 concentration prediction is of great significance to environmental protection and human health.Achieving accurate prediction of PM2.5 concentration has become an important research task.However,PM2.5 pollutants can spread in the earth’s atmosphere,causing mutual influence between different cities.To effectively capture the air pollution relationship between cities,this paper proposes a novel spatiotemporal model combining graph attention neural network(GAT)and gated recurrent unit(GRU),named GAT-GRU for PM2.5 concentration prediction.Specifically,GAT is used to learn the spatial dependence of PM2.5 concentration data in different cities,and GRU is to extract the temporal dependence of the long-term data series.The proposed model integrates the learned spatio-temporal dependencies to capture long-term complex spatio-temporal features.Considering that air pollution is related to the meteorological conditions of the city,the knowledge acquired from meteorological data is used in the model to enhance PM2.5 prediction performance.The input of the GAT-GRU model consists of PM2.5 concentration data and meteorological data.In order to verify the effectiveness of the proposed GAT-GRU prediction model,this paper designs experiments on real-world datasets compared with other baselines.Experimental results prove that our model achieves excellent performance in PM2.5 concentration prediction.
文摘Numerous works prove that existing neighbor-averaging graph neural networks(GNNs)cannot efficiently catch structure features,and many works show that injecting structure,distance,position,or spatial features can significantly improve the performance of GNNs,however,injecting high-level structure and distance into GNNs is an intuitive but untouched idea.This work sheds light on this issue and proposes a scheme to enhance graph attention networks(GATs)by encoding distance and hop-wise structure statistics.Firstly,the hop-wise structure and distributional distance information are extracted based on several hop-wise ego-nets of every target node.Secondly,the derived structure information,distance information,and intrinsic features are encoded into the same vector space and then added together to get initial embedding vectors.Thirdly,the derived embedding vectors are fed into GATs,such as GAT and adaptive graph diffusion network(AGDN)to get the soft labels.Fourthly,the soft labels are fed into correct and smooth(C&S)to conduct label propagation and get final predictions.Experiments show that the distance and hop-wise structures encoding enhanced graph attention networks(DHSEGATs)achieve a competitive result.
基金the National Natural Science Foundation of China(No.61461027,61762059)the Provincial Science and Technology Program supported the Key Project of Natural Science Foundation of Gansu Province(No.22JR5RA226)。
文摘Considering the nonlinear structure and spatial-temporal correlation of traffic network,and the influence of potential correlation between nodes of traffic network on the spatial features,this paper proposes a traffic speed prediction model based on the combination of graph attention network with self-adaptive adjacency matrix(SAdpGAT)and bidirectional gated recurrent unit(BiGRU).First-ly,the model introduces graph attention network(GAT)to extract the spatial features of real road network and potential road network respectively in spatial dimension.Secondly,the spatial features are input into BiGRU to extract the time series features.Finally,the prediction results of the real road network and the potential road network are connected to generate the final prediction results of the model.The experimental results show that the prediction accuracy of the proposed model is im-proved obviously on METR-LA and PEMS-BAY datasets,which proves the advantages of the pro-posed spatial-temporal model in traffic speed prediction.
基金supported in part by the Science and Technology Innovation 2030-“New Generation of Artificial Intelligence”Major Project(No.2021ZD0111000)Henan Provincial Science and Technology Research Project(No.232102211039).
文摘The growing prevalence of knowledge reasoning using knowledge graphs(KGs)has substantially improved the accuracy and efficiency of intelligent medical diagnosis.However,current models primarily integrate electronic medical records(EMRs)and KGs into the knowledge reasoning process,ignoring the differing significance of various types of knowledge in EMRs and the diverse data types present in the text.To better integrate EMR text information,we propose a novel intelligent diagnostic model named the Graph ATtention network incorporating Text representation in knowledge reasoning(GATiT),which comprises text representation,subgraph construction,knowledge reasoning,and diagnostic classification.In the text representation process,GATiT uses a pre-trained model to obtain text representations of the EMRs and additionally enhances embeddings by including chief complaint information and numerical information in the input.In the subgraph construction process,GATiT constructs text subgraphs and disease subgraphs from the KG,utilizing EMR text and the disease to be diagnosed.To differentiate the varying importance of nodes within the subgraphs features such as node categories,relevance scores,and other relevant factors are introduced into the text subgraph.Themessage-passing strategy and attention weight calculation of the graph attention network are adjusted to learn these features in the knowledge reasoning process.Finally,in the diagnostic classification process,the interactive attention-based fusion method integrates the results of knowledge reasoning with text representations to produce the final diagnosis results.Experimental results on multi-label and single-label EMR datasets demonstrate the model’s superiority over several state-of-theart methods.
基金supported byNationalNatural Science Foundation of China(52274205)and Project of Education Department of Liaoning Province(LJKZ0338).
文摘Automatic text summarization(ATS)plays a significant role in Natural Language Processing(NLP).Abstractive summarization produces summaries by identifying and compressing the most important information in a document.However,there are only relatively several comprehensively evaluated abstractive summarization models that work well for specific types of reports due to their unstructured and oral language text characteristics.In particular,Chinese complaint reports,generated by urban complainers and collected by government employees,describe existing resident problems in daily life.Meanwhile,the reflected problems are required to respond speedily.Therefore,automatic summarization tasks for these reports have been developed.However,similar to traditional summarization models,the generated summaries still exist problems of informativeness and conciseness.To address these issues and generate suitably informative and less redundant summaries,a topic-based abstractive summarization method is proposed to obtain global and local features.Additionally,a heterogeneous graph of the original document is constructed using word-level and topic-level features.Experiments and analyses on public review datasets(Yelp and Amazon)and our constructed dataset(Chinese complaint reports)show that the proposed framework effectively improves the performance of the abstractive summarization model for Chinese complaint reports.
基金This work is supposed by the Science and Technology Projects of China Southern Power Grid(YNKJXM20222402).
文摘Advanced carbon emission factors of a power grid can provide users with effective carbon reduction advice,which is of immense importance in mobilizing the entire society to reduce carbon emissions.The method of calculating node carbon emission factors based on the carbon emissions flow theory requires real-time parameters of a power grid.Therefore,it cannot provide carbon factor information beforehand.To address this issue,a prediction model based on the graph attention network is proposed.The model uses a graph structure that is suitable for the topology of the power grid and designs a supervised network using the loads of the grid nodes and the corresponding carbon factor data.The network extracts features and transmits information more suitable for the power system and can flexibly adjust the equivalent topology,thereby increasing the diversity of the structure.Its input and output data are simple,without the power grid parameters.We demonstrated its effect by testing IEEE-39 bus and IEEE-118 bus systems with average error rates of 2.46%and 2.51%.
基金supported by China Southern Power Grid Co.,Ltd.Yunnan Electric Power Dispatching Control Center(Cyber-physical-based“Source-network-load-storage”Coordinated Dispatch and Control Technologies and Application System Development,sub-project YNKJXM20222463)。
文摘The high proportion of renewable energy integration and the dynamic changes in grid topology necessitate the enhancement of voltage/var control(VVC)to manage voltage fluctuations more rapidly.Traditional model-based control algorithms are becoming increasingly incompetent for VVC due to their high model dependence and slow online computation speed.To alleviate these issues,this paper introduces a graph attention network(GAT)based deep reinforcement learning for VVC of topologically variable power system.Firstly,combining the physical information of the actual power grid,a physics-informed GAT is proposed and embedded into the proximal policy optimization(PPO)algorithm.The GAT-PPO algorithm can capture topological and spatial correlations among the node features to tackle topology changes.To address the slow training,the Relief F-S algorithm identifies critical state variables,significantly reducing the dimensionality of state space.Then,the training samples retained in the experience buffer are designed to mitigate the sparse reward issue.Finally,the validation on the modified IEEE 39-bus system and an actual power grid demonstrates superior performance of the proposed algorithm compared with state-of-the-art algorithms,including PPO algorithm and twin delayed deep deterministic policy gradient(TD3)algorithm.The proposed algorithm exhibits enhanced convergence during training,faster solution speed,and improved VVC performance,even in scenarios involving grid topology changes and increased renewable energy integration.Meanwhile,in the adopted cases,the network loss is reduced by 6.9%,10.8%,and 7.7%,respectively,demonstrating favorable economic outcomes.
基金supported by the National Natural Science Foundation of China(grant 62136004 to D.Z.,grant 62272226 to W.S.,and grant 82173078 to J.S.)the National Key Research and Development Program of China(grant 2019YFA0803000 to J.S.)+1 种基金the Excellent Youth Foundation of Zhejiang Scientific(grant R22H1610037 to J.S.)the Zhejiang Provincial Natural Science Foundation(grant 2022C03037 to J.S.).
文摘Imaging mass cytometry(IMC)enables the high-resolution spatial profiling of tumor microenvironment,but its clinical utility for prospective prediction remains underdeveloped.In this study,we integrated IMC into a clinical trial of hepatocellular carcinoma(HCC)patients undergoing combination therapy with programmed death-1 blockade and transarterial chemoembolization.We analyzed 281 regions of interest from 43 patients using a custom 40-marker IMC panel and developed a novel superpixel-based graph attention network,IMCSGAT,to model spatial cell interactions within the tumor microenvironment.IMCSGAT enabled accurate multitask prediction of key clinical features,including Barcelona Clinic Liver Cancer stage,trabecular histologic subtype,and treatment response.Compared to state-of-the-art methods,IMCSGAT achieved superior performance across all classification tasks.Spatial interaction analysis revealed that resident macrophage-centered interactions,particularly those with NK and T cells,were enriched in responders and predictive of therapeutic outcome.These findings were validated in a murine HCC model,reinforcing the role of innate immune architecture in shaping the treatment response.This study establishes IMCSGAT as a powerful spatial learning framework for high-dimensional IMC data,with potential applications in clinical outcome prediction and personalized therapy design for HCC.Our results provide a blueprint for the broader use of spatial analytics in precision oncology.
基金supported by Research and Application of Edge IoT Technology for Distributed New Energy Consumption in Distribution Areas,Project Number(5108-202218280A-2-394-XG)。
文摘For permanent faults(PF)in the power communication network(PCN),such as link interruptions,the timesensitive networking(TSN)relied on by PCN,typically employs spatial redundancy fault-tolerance methods to keep service stability and reliability,which often limits TSN scheduling performance in fault-free ideal states.So this paper proposes a graph attention residual network-based routing and fault-tolerant scheduling mechanism(GRFS)for data flow in PCN,which specifically includes a communication system architecture for integrated terminals based on a cyclic queuing and forwarding(CQF)model and fault recovery method,which reduces the impact of faults by simplified scheduling configurations of CQF and fault-tolerance of prioritizing the rerouting of faulty time-sensitive(TS)flows;considering that PF leading to changes in network topology is more appropriately solved by doing routing and time slot injection decisions hop-by-hop,and that reasonable network load can reduce the damage caused by PF and reserve resources for the rerouting of faulty TS flows,an optimization model for joint routing and scheduling is constructed with scheduling success rate as the objective,and with traffic latency and network load as constraints;to catch changes in TSN topology and traffic load,a D3QN algorithm based on a multi-head graph attention residual network(MGAR)is designed to solve the problem model,where the MGAR based encoder reconstructs the TSN status into feature embedding vectors,and a dueling network decoder performs decoding tasks on the reconstructed feature embedding vectors.Simulation results show that GRFS outperforms heuristic fault-tolerance algorithms and other benchmark schemes by approximately 10%in routing and scheduling success rate in ideal states and 5%in rerouting and rescheduling success rate in fault states.
基金supported by the MSIT(Ministry of Science and ICT),Republic of Korea,under the ITRC(Information Technology Research Centre)support program(IITP-2024-RS-2024-00437191)supervised by the IITP(Institute for Information&Communications Technology Planning&Evaluation).
文摘Dialogue State Tracking(DST)is a critical component of task-oriented spoken dialogue systems(SDS),tasked with maintaining an accurate representation of the conversational state by predicting slots and their corresponding values.Recent advances leverage Large Language Models(LLMs)with prompt-based tuning to improve tracking accuracy and efficiency.However,these approaches often incur substantial computational and memory overheads and typically address slot extraction implicitly within prompts,without explicitly modeling the complex dependencies between slots and values.In this work,we propose PUGG,a novel DST framework that constructs schema-driven prompts to fine-tune GPT-2 and utilizes its tokenizer to implement a memory encoder.PUGG explicitly extracts slot values via GPT-2 and employs Graph Attention Networks(GATs)to model and reason over the intricate relationships between slots and their associated values.We evaluate PUGG on four publicly available datasets,where it achieves stateof-the-art performance across multiple evaluation metrics,highlighting its robustness and generalizability in diverse conversational scenarios.Our results indicate that the integration of GPT-2 substantially reduces model complexity and memory consumption by streamlining key processes.Moreover,prompt tuning enhances the model’s flexibility and precision in extracting relevant slot-value pairs,while the incorporation of GATs facilitates effective relational reasoning,leading to improved dialogue state representations.
文摘Most existing knowledge graph relationship prediction methods are unable to capture the complex information of multi-relational knowledge graphs,thus overlooking key details contained in different entity pairs and making it difficult to aggregate more complex relational features.Moreover,the insufficient capture of multi-hop relational information limits the processing capability of the global structure of the graph and reduces the accuracy of the knowledge graph completion task.This paper uses graph neural networks to construct new message functions for different relations,which can be defined as the rotation from the source entity to the target entity in the complex vector space for each relation,thereby improving the relation perception.To further enrich the relational diversity of different entities,we capture themulti-hop structural information in complex graph structure relations by incorporating two-hop relations for each entity and adding auxiliary edges to various relation combinations in the knowledge graph,thereby aggregating more complex relations and improving the reasoning ability of complex relational information.To verify the effectiveness of the proposed method,we conducted experiments on the WN18RR and FB15k-237 standard datasets.The results show that the method proposed in this study outperforms most existing methods.
基金Project supported by the Natural Science Foundation of Shandong Province,China(Grant No.ZR2021MF049)the Joint Fund of the Natural Science Foundation of Shandong Province,China(Grant Nos.ZR2022LLZ012 and ZR2021LLZ001)the Key Research and Development Program of Shandong Province,China(Grant No.2023CXGC010901)。
文摘Quantum error-correcting codes are essential for fault-tolerant quantum computing,as they effectively detect and correct noise-induced errors by distributing information across multiple physical qubits.The subsystem surface code with three-qubit check operators demonstrates significant application potential due to its simplified measurement operations and low logical error rates.However,the existing minimum-weight perfect matching(MWPM)algorithm exhibits high computational complexity and lacks flexibility in large-scale systems.Therefore,this paper proposes a decoder based on a graph attention network(GAT),representing error syndromes as undirected graphs with edge weights,and employing a multihead attention mechanism to efficiently aggregate node features and enable parallel computation.Compared to MWPM,the GAT decoder exhibits linear growth in computational complexity,adapts to different quantum code structures,and demonstrates stronger robustness under high physical error rates.The experimental results demonstrate that the proposed decoder achieves an overall accuracy of 89.95%under various small code lattice sizes(L=2,3,4,5),with the logical error rate threshold increasing to 0.0078,representing an improvement of approximately 13.04%compared to the MWPM decoder.This result significantly outperforms traditional methods,showcasing superior performance under small code lattice sizes and providing a more efficient decoding solution for large-scale quantum error correction.
基金the National Natural Science Foundation of China(Grant Nos.11804076 and 91961204)the Fundamental Research Funds for the Central Universities of China(No.B210202151)the Changzhou Science and Technology Plan(No.CZ520012712).
文摘Due to coexistence of huge number of structural isomers,global search for the ground-state structures of atomic clusters is a challenging issue.The difficulty also originates from the computational cost of ab initio methods for describing the potential energy surface.Recently,machine learning techniques have been widely utilized to accelerate materials discovery and molecular simulation.Compared to the commonly used artificial neural network,graph network is naturally suitable for clusters with flexible geometric environment of each atom.Herein we develop a cluster graph attention network(CGANet)by aggregating information of neighboring vertices and edges using attention mechanism,which can precisely predict the binding energy and force of silver clusters with root mean square error of 5.4 meV/atom and mean absolute error of 42.3 meV/Å,respectively.As a proof-of-concept,we have performed global optimization of mediumsized Agn clusters(n=14–26)by combining CGANet and genetic algorithm.The reported ground-state structures for n=14–21,have been successfully reproduced,while entirely new lowest-energy structures are obtained for n=22–26.In addition to the description of potential energy surface,the CGANet is also applied to predict the electronic properties of clusters,such as HOMO energy and HOMO-LUMO gap.With accuracy comparable to ab initio methods and acceleration by at least two orders of magnitude,CGANet holds great promise in global search of lowest-energy structures of large clusters and inverse design of functional clusters.
基金This work was supported by the National Key Research and Development Program of China under Grant No.2021YFB3100600the Youth Innovation Promotion Association of Chinese Academy of Sciences under Grant No.2021153the State Key Program of National Natural Science Foundation of China under Grant No.U2336202.
文摘Event detection(ED)seeks to recognize event triggers and classify them into the predefined event types.Chinese ED is formulated as a character-level task owing to the uncertain word boundaries.Prior methods try to incorpo-rate word-level information into characters to enhance their semantics.However,they experience two problems.First,they fail to incorporate word-level information into each character the word encompasses,causing the insufficient word-charac-ter interaction problem.Second,they struggle to distinguish events of similar types with limited annotated instances,which is called the event confusing problem.This paper proposes a novel model named Label-Aware Heterogeneous Graph Attention Network(L-HGAT)to address these two problems.Specifically,we first build a heterogeneous graph of two node types and three edge types to maximally preserve word-character interactions,and then deploy a heterogeneous graph attention network to enhance the semantic propagation between characters and words.Furthermore,we design a pushing-away game to enlarge the predicting gap between the ground-truth event type and its confusing counterpart for each character.Experimental results show that our L-HGAT model consistently achieves superior performance over prior competitive methods.
