With the escalating complexity in production scenarios, vast amounts of production information are retained within enterprises in the industrial domain. Probing questions of how to meticulously excavate value from com...With the escalating complexity in production scenarios, vast amounts of production information are retained within enterprises in the industrial domain. Probing questions of how to meticulously excavate value from complex document information and establish coherent information links arise. In this work, we present a framework for knowledge graph construction in the industrial domain, predicated on knowledge-enhanced document-level entity and relation extraction. This approach alleviates the shortage of annotated data in the industrial domain and models the interplay of industrial documents. To augment the accuracy of named entity recognition, domain-specific knowledge is incorporated into the initialization of the word embedding matrix within the bidirectional long short-term memory conditional random field (BiLSTM-CRF) framework. For relation extraction, this paper introduces the knowledge-enhanced graph inference (KEGI) network, a pioneering method designed for long paragraphs in the industrial domain. This method discerns intricate interactions among entities by constructing a document graph and innovatively integrates knowledge representation into both node construction and path inference through TransR. On the application stratum, BiLSTM-CRF and KEGI are utilized to craft a knowledge graph from a knowledge representation model and Chinese fault reports for a steel production line, specifically SPOnto and SPFRDoc. The F1 value for entity and relation extraction has been enhanced by 2% to 6%. The quality of the extracted knowledge graph complies with the requirements of real-world production environment applications. The results demonstrate that KEGI can profoundly delve into production reports, extracting a wealth of knowledge and patterns, thereby providing a comprehensive solution for production management.展开更多
In recent years,deep learning methods have developed rapidly and found application in many fields,including natural language processing.In the field of aspect-level sentiment analysis,deep learning methods can also gr...In recent years,deep learning methods have developed rapidly and found application in many fields,including natural language processing.In the field of aspect-level sentiment analysis,deep learning methods can also greatly improve the performance of models.However,previous studies did not take into account the relationship between user feature extraction and contextual terms.To address this issue,we use data feature extraction and deep learning combined to develop an aspect-level sentiment analysis method.To be specific,we design user comment feature extraction(UCFE)to distill salient features from users’historical comments and transform them into representative user feature vectors.Then,the aspect-sentence graph convolutional neural network(ASGCN)is used to incorporate innovative techniques for calculating adjacency matrices;meanwhile,ASGCN emphasizes capturing nuanced semantics within relationships among aspect words and syntactic dependency types.Afterward,three embedding methods are devised to embed the user feature vector into the ASGCN model.The empirical validations verify the effectiveness of these models,consistently surpassing conventional benchmarks and reaffirming the indispensable role of deep learning in advancing sentiment analysis methodologies.展开更多
Let BH,K = {BH,K(t), t ∈ R+} be a bifractional Brownian motion in Rd. This process is a selfsimilar Gaussian process depending on two parameters H and K and it constitutes a natural generalization of fractional Brown...Let BH,K = {BH,K(t), t ∈ R+} be a bifractional Brownian motion in Rd. This process is a selfsimilar Gaussian process depending on two parameters H and K and it constitutes a natural generalization of fractional Brownian motion (which is obtained for K = 1). The exact Hausdorff measures of the image, graph and the level set of BH,K are investigated. The results extend the corresponding results proved by Talagrand and Xiao for fractional Brownian motion.展开更多
A novel approach for assessing the robustness of an equilibrium in conflict resolution is presented. Roughly, an equilibrium is robust if it is resilient, or resistant to deviation. Robustness assessment is based on a...A novel approach for assessing the robustness of an equilibrium in conflict resolution is presented. Roughly, an equilibrium is robust if it is resilient, or resistant to deviation. Robustness assessment is based on a new concept called Level of Freedom, which evaluates the relative freedom of a decision maker to escape an equilibrium. Resolutions of a conflict can be affected by changes in decision makers＇ preferences, which may destabilize an equilibrium, causing the conflict to evolve. Hence, a conflict may become long-term and thereby continue to evolve, even after reaching an equilibrium. The new robustness measure is used to rank equilibria based on robustness, to facilitate distinguishing equiiibria that are relatively sustainable. An absolutely robust equilibrium is a special case in which the level of freedom is at an absolute minimum for each individual stability definition.展开更多
Partitioning is a fundamental problem with applications to many areas including data mining, parellel processing and Very-large-scale integration （VLSI） design. An effective multi-level algorithm for bisecting graph...Partitioning is a fundamental problem with applications to many areas including data mining, parellel processing and Very-large-scale integration （VLSI） design. An effective multi-level algorithm for bisecting graph is proposed. During its coarsening phase, an improved matching approach based on the global information of the graph core is developed with its guidance function. During the refinement phase, the vertex gain is exploited as ant＇s heuristic information and a positive feedback method based on pheromone trails is used to find the global approximate bipartitioning. It is implemented with American National Standards Institute （ANSI） C and compared to MeTiS. The experimental evaluation shows that it performs well and produces encouraging solutions on 18 different graphs benchmarks.展开更多
Traffic scene captioning technology automatically generates one or more sentences to describe the content of traffic scenes by analyzing the content of the input traffic scene images,ensuring road safety while providi...Traffic scene captioning technology automatically generates one or more sentences to describe the content of traffic scenes by analyzing the content of the input traffic scene images,ensuring road safety while providing an important decision-making function for sustainable transportation.In order to provide a comprehensive and reasonable description of complex traffic scenes,a traffic scene semantic captioningmodel withmulti-stage feature enhancement is proposed in this paper.In general,the model follows an encoder-decoder structure.First,multilevel granularity visual features are used for feature enhancement during the encoding process,which enables the model to learn more detailed content in the traffic scene image.Second,the scene knowledge graph is applied to the decoding process,and the semantic features provided by the scene knowledge graph are used to enhance the features learned by the decoder again,so that themodel can learn the attributes of objects in the traffic scene and the relationships between objects to generate more reasonable captions.This paper reports extensive experiments on the challenging MS-COCO dataset,evaluated by five standard automatic evaluation metrics,and the results show that the proposed model has improved significantly in all metrics compared with the state-of-the-art methods,especially achieving a score of 129.0 on the CIDEr-D evaluation metric,which also indicates that the proposed model can effectively provide a more reasonable and comprehensive description of the traffic scene.展开更多
The reconstruction of high-resolution sea-level variation curves in deep time based on the standard car-bonate microfacies knowledge graph(SMFKG)is of great scientific significance for exploring the Earth system evolu...The reconstruction of high-resolution sea-level variation curves in deep time based on the standard car-bonate microfacies knowledge graph(SMFKG)is of great scientific significance for exploring the Earth system evolution and predicting future sea-level and climate changes.In this study,the concepts,attri-butes,and relationships among standard carbonate microfacies(SMF)are comprehensively analyzed;an ontology layer is established and its data layer is constructed using thin-section descriptions;and finally,the SMFKG is established.Additionally,based on the knowledge graph,an application for automatically identifying SMF using identification markers and reconstructing the high-resolution relative sea-level variation curve using the SMF and facies zones is compiled.Then,all thin sections of the late Ediacaran Dengying Formation in the western margin of the Yangtze Platform are observed and described in detail,the SMF and facies zones are identified automatically,and the relative sea-level curve is recon-structed automatically using the SMFKG.The reconstruction results show that the Yangtze Platform experienced four sea-level rise and fall cycles in the late Ediacaran,of which two intense regressions led to subaerial-exposed unconformities in the interior and top of the Dengying Formation,which is highly consistent with previous research results.This shows that the high-resolution relative sea-level variation curve in deep time can be reconstructed efficiently and intelligently using the SMFKG.Additionally,in the near future,the combination of an automatic digital slide-scanning system,machine-learning techniques,and the SMFKG can achieve one-stop fully automatic SMF recognition and reconstruction of high-resolution relative sea-level variation curves in deep time,which has a high application value.展开更多
Traffic flow prediction plays an important role in intelligent transportation systems and is of great significance in the applications of traffic control and urban planning.Due to the complexity of road traffic flow d...Traffic flow prediction plays an important role in intelligent transportation systems and is of great significance in the applications of traffic control and urban planning.Due to the complexity of road traffic flow data,traffic flow prediction has been one of the challenging tasks to fully exploit the spatiotemporal characteristics of roads to improve prediction accuracy.In this study,a combined flow direction level traffic flow prediction graph convolutional network(GCN)and long short-term memory(LSTM)model based on spatiotemporal characteristics is proposed.First,a GCN model is employed to capture the topological structure of the data graph and extract the spatial features of road networks.Additionally,due to the capability to handle long-term dependencies,the longterm memory is used to predict the time series of traffic flow and extract the time features.The proposed model is evaluated using real-world data,which are obtained from the intersection of Liuquan Road and Zhongrun Avenue in the Zibo High-Tech Zone of China.The results show that the developed combined GCNLSTM flow direction level traffic flow prediction model can perform better than the single models of the LSTM model and GCN model,and the combined ARIMA-LSTM model in traffic flow has a strong spatiotemporal correlation.展开更多
The introduction of the social networking platform has drastically affected the way individuals interact. Even though most of the effects have been positive, there exist some serious threats associated with the intera...The introduction of the social networking platform has drastically affected the way individuals interact. Even though most of the effects have been positive, there exist some serious threats associated with the interactions on a social networking website. A considerable proportion of the crimes that occur are initiated through a social networking platform [1]. Almost 33% of the crimes on the internet are initiated through a social networking website [1]. Moreover activities like spam messages create unnecessary traffic and might affect the user base of a social networking platform. As a result preventing interactions with malicious intent and spam activities becomes crucial. This work attempts to detect the same in a social networking platform by considering a social network as a weighted graph wherein each node, which represents an individual in the social network, stores activities of other nodes with respect to itself in an optimized format which is referred to as localized data set. The weights associated with the edges in the graph represent the trust relationship between profiles. The weights of the edges along with the localized data set are used to infer whether nodes in the social network are compromised and are performing spam or malicious activities.展开更多
Let A be an M-matrix and B be a Z-matrix. In this paper we reveal the spectral relationship of A and B under some interesting conditions. Applying this result, we solve an open problem on splittings of an M-matrix and...Let A be an M-matrix and B be a Z-matrix. In this paper we reveal the spectral relationship of A and B under some interesting conditions. Applying this result, we solve an open problem on splittings of an M-matrix and partially answer an open problem on the level diagrams for A and B.展开更多
Segmentation of three-dimensional(3D) complicated structures is of great importance for many real applications.In this work we combine graph cut minimization method with a variant of the level set idea for 3D segmenta...Segmentation of three-dimensional(3D) complicated structures is of great importance for many real applications.In this work we combine graph cut minimization method with a variant of the level set idea for 3D segmentation based on the Mumford-Shah model.Compared with the traditional approach for solving the Euler-Lagrange equation we do not need to solve any partial differential equations.Instead,the minimum cut on a special designed graph need to be computed.The method is tested on data with complicated structures.It is rather stable with respect to initial value and the algorithm is nearly parameter free.Experiments show that it can solve large problems much faster than traditional approaches.展开更多
基金supported by the National Science and Technology Innovation 2030 New Generation Artificial Intelligence Major Project(Grant No.2018AAA0101800)the National Natural Science Foundation of China(Grant No.72271188).
文摘With the escalating complexity in production scenarios, vast amounts of production information are retained within enterprises in the industrial domain. Probing questions of how to meticulously excavate value from complex document information and establish coherent information links arise. In this work, we present a framework for knowledge graph construction in the industrial domain, predicated on knowledge-enhanced document-level entity and relation extraction. This approach alleviates the shortage of annotated data in the industrial domain and models the interplay of industrial documents. To augment the accuracy of named entity recognition, domain-specific knowledge is incorporated into the initialization of the word embedding matrix within the bidirectional long short-term memory conditional random field (BiLSTM-CRF) framework. For relation extraction, this paper introduces the knowledge-enhanced graph inference (KEGI) network, a pioneering method designed for long paragraphs in the industrial domain. This method discerns intricate interactions among entities by constructing a document graph and innovatively integrates knowledge representation into both node construction and path inference through TransR. On the application stratum, BiLSTM-CRF and KEGI are utilized to craft a knowledge graph from a knowledge representation model and Chinese fault reports for a steel production line, specifically SPOnto and SPFRDoc. The F1 value for entity and relation extraction has been enhanced by 2% to 6%. The quality of the extracted knowledge graph complies with the requirements of real-world production environment applications. The results demonstrate that KEGI can profoundly delve into production reports, extracting a wealth of knowledge and patterns, thereby providing a comprehensive solution for production management.
基金This work is partly supported by the Fundamental Research Funds for the Central Universities(CUC230A013)It is partly supported by Natural Science Foundation of Beijing Municipality(No.4222038)It is also supported by National Natural Science Foundation of China(Grant No.62176240).
文摘In recent years,deep learning methods have developed rapidly and found application in many fields,including natural language processing.In the field of aspect-level sentiment analysis,deep learning methods can also greatly improve the performance of models.However,previous studies did not take into account the relationship between user feature extraction and contextual terms.To address this issue,we use data feature extraction and deep learning combined to develop an aspect-level sentiment analysis method.To be specific,we design user comment feature extraction(UCFE)to distill salient features from users’historical comments and transform them into representative user feature vectors.Then,the aspect-sentence graph convolutional neural network(ASGCN)is used to incorporate innovative techniques for calculating adjacency matrices;meanwhile,ASGCN emphasizes capturing nuanced semantics within relationships among aspect words and syntactic dependency types.Afterward,three embedding methods are devised to embed the user feature vector into the ASGCN model.The empirical validations verify the effectiveness of these models,consistently surpassing conventional benchmarks and reaffirming the indispensable role of deep learning in advancing sentiment analysis methodologies.
基金supported by National Natural Science Foundation of China (Grant No.10721091)
文摘Let BH,K = {BH,K(t), t ∈ R+} be a bifractional Brownian motion in Rd. This process is a selfsimilar Gaussian process depending on two parameters H and K and it constitutes a natural generalization of fractional Brownian motion (which is obtained for K = 1). The exact Hausdorff measures of the image, graph and the level set of BH,K are investigated. The results extend the corresponding results proved by Talagrand and Xiao for fractional Brownian motion.
文摘A novel approach for assessing the robustness of an equilibrium in conflict resolution is presented. Roughly, an equilibrium is robust if it is resilient, or resistant to deviation. Robustness assessment is based on a new concept called Level of Freedom, which evaluates the relative freedom of a decision maker to escape an equilibrium. Resolutions of a conflict can be affected by changes in decision makers＇ preferences, which may destabilize an equilibrium, causing the conflict to evolve. Hence, a conflict may become long-term and thereby continue to evolve, even after reaching an equilibrium. The new robustness measure is used to rank equilibria based on robustness, to facilitate distinguishing equiiibria that are relatively sustainable. An absolutely robust equilibrium is a special case in which the level of freedom is at an absolute minimum for each individual stability definition.
基金the International Cooperation Project of Ministry of Science and Technology of P. R. China (GrantNo.CB7-2-01)SEC E-Institute: Shanghai High Institutions Grid
文摘Partitioning is a fundamental problem with applications to many areas including data mining, parellel processing and Very-large-scale integration （VLSI） design. An effective multi-level algorithm for bisecting graph is proposed. During its coarsening phase, an improved matching approach based on the global information of the graph core is developed with its guidance function. During the refinement phase, the vertex gain is exploited as ant＇s heuristic information and a positive feedback method based on pheromone trails is used to find the global approximate bipartitioning. It is implemented with American National Standards Institute （ANSI） C and compared to MeTiS. The experimental evaluation shows that it performs well and produces encouraging solutions on 18 different graphs benchmarks.
基金funded by(i)Natural Science Foundation China(NSFC)under Grant Nos.61402397,61263043,61562093 and 61663046(ii)Open Foundation of Key Laboratory in Software Engineering of Yunnan Province:No.2020SE304.(iii)Practical Innovation Project of Yunnan University,Project Nos.2021z34,2021y128 and 2021y129.
文摘Traffic scene captioning technology automatically generates one or more sentences to describe the content of traffic scenes by analyzing the content of the input traffic scene images,ensuring road safety while providing an important decision-making function for sustainable transportation.In order to provide a comprehensive and reasonable description of complex traffic scenes,a traffic scene semantic captioningmodel withmulti-stage feature enhancement is proposed in this paper.In general,the model follows an encoder-decoder structure.First,multilevel granularity visual features are used for feature enhancement during the encoding process,which enables the model to learn more detailed content in the traffic scene image.Second,the scene knowledge graph is applied to the decoding process,and the semantic features provided by the scene knowledge graph are used to enhance the features learned by the decoder again,so that themodel can learn the attributes of objects in the traffic scene and the relationships between objects to generate more reasonable captions.This paper reports extensive experiments on the challenging MS-COCO dataset,evaluated by five standard automatic evaluation metrics,and the results show that the proposed model has improved significantly in all metrics compared with the state-of-the-art methods,especially achieving a score of 129.0 on the CIDEr-D evaluation metric,which also indicates that the proposed model can effectively provide a more reasonable and comprehensive description of the traffic scene.
基金supported by the IUGS Deep-time Digital Earth(DDE)Big Science Program,National Natural Science Foundation of China(No.42050104,No.42102138 and No.U19B6003)the Open Fund(DGERA20221103)of Key Laboratory of Deep-time Geography and Environment Reconstruction and Applications of Ministry of Natural ResourcesChengdu University of Technology,China and the Open Fund(PLC20210202)of the State Key Labora-tory of Oil and Gas Reservoir Geology and Exploitation(Chengdu University of Technology,China).
文摘The reconstruction of high-resolution sea-level variation curves in deep time based on the standard car-bonate microfacies knowledge graph(SMFKG)is of great scientific significance for exploring the Earth system evolution and predicting future sea-level and climate changes.In this study,the concepts,attri-butes,and relationships among standard carbonate microfacies(SMF)are comprehensively analyzed;an ontology layer is established and its data layer is constructed using thin-section descriptions;and finally,the SMFKG is established.Additionally,based on the knowledge graph,an application for automatically identifying SMF using identification markers and reconstructing the high-resolution relative sea-level variation curve using the SMF and facies zones is compiled.Then,all thin sections of the late Ediacaran Dengying Formation in the western margin of the Yangtze Platform are observed and described in detail,the SMF and facies zones are identified automatically,and the relative sea-level curve is recon-structed automatically using the SMFKG.The reconstruction results show that the Yangtze Platform experienced four sea-level rise and fall cycles in the late Ediacaran,of which two intense regressions led to subaerial-exposed unconformities in the interior and top of the Dengying Formation,which is highly consistent with previous research results.This shows that the high-resolution relative sea-level variation curve in deep time can be reconstructed efficiently and intelligently using the SMFKG.Additionally,in the near future,the combination of an automatic digital slide-scanning system,machine-learning techniques,and the SMFKG can achieve one-stop fully automatic SMF recognition and reconstruction of high-resolution relative sea-level variation curves in deep time,which has a high application value.
基金supported by the National Natural Science Foundation of China (Grant Nos.71901134&51878165)the National Science Foundation for Distinguished Young Scholars (Grant No.51925801).
文摘Traffic flow prediction plays an important role in intelligent transportation systems and is of great significance in the applications of traffic control and urban planning.Due to the complexity of road traffic flow data,traffic flow prediction has been one of the challenging tasks to fully exploit the spatiotemporal characteristics of roads to improve prediction accuracy.In this study,a combined flow direction level traffic flow prediction graph convolutional network(GCN)and long short-term memory(LSTM)model based on spatiotemporal characteristics is proposed.First,a GCN model is employed to capture the topological structure of the data graph and extract the spatial features of road networks.Additionally,due to the capability to handle long-term dependencies,the longterm memory is used to predict the time series of traffic flow and extract the time features.The proposed model is evaluated using real-world data,which are obtained from the intersection of Liuquan Road and Zhongrun Avenue in the Zibo High-Tech Zone of China.The results show that the developed combined GCNLSTM flow direction level traffic flow prediction model can perform better than the single models of the LSTM model and GCN model,and the combined ARIMA-LSTM model in traffic flow has a strong spatiotemporal correlation.
文摘The introduction of the social networking platform has drastically affected the way individuals interact. Even though most of the effects have been positive, there exist some serious threats associated with the interactions on a social networking website. A considerable proportion of the crimes that occur are initiated through a social networking platform [1]. Almost 33% of the crimes on the internet are initiated through a social networking website [1]. Moreover activities like spam messages create unnecessary traffic and might affect the user base of a social networking platform. As a result preventing interactions with malicious intent and spam activities becomes crucial. This work attempts to detect the same in a social networking platform by considering a social network as a weighted graph wherein each node, which represents an individual in the social network, stores activities of other nodes with respect to itself in an optimized format which is referred to as localized data set. The weights associated with the edges in the graph represent the trust relationship between profiles. The weights of the edges along with the localized data set are used to infer whether nodes in the social network are compromised and are performing spam or malicious activities.
文摘Let A be an M-matrix and B be a Z-matrix. In this paper we reveal the spectral relationship of A and B under some interesting conditions. Applying this result, we solve an open problem on splittings of an M-matrix and partially answer an open problem on the level diagrams for A and B.
基金support from the Centre for Integrated Petroleum Research(CIPR),University of Bergen, Norway,and Singapore MOE Grant T207B2202NRF2007IDMIDM002-010
文摘Segmentation of three-dimensional(3D) complicated structures is of great importance for many real applications.In this work we combine graph cut minimization method with a variant of the level set idea for 3D segmentation based on the Mumford-Shah model.Compared with the traditional approach for solving the Euler-Lagrange equation we do not need to solve any partial differential equations.Instead,the minimum cut on a special designed graph need to be computed.The method is tested on data with complicated structures.It is rather stable with respect to initial value and the algorithm is nearly parameter free.Experiments show that it can solve large problems much faster than traditional approaches.
