Multi-modal knowledge graph completion(MMKGC)aims to complete missing entities or relations in multi-modal knowledge graphs,thereby discovering more previously unknown triples.Due to the continuous growth of data and ...Multi-modal knowledge graph completion(MMKGC)aims to complete missing entities or relations in multi-modal knowledge graphs,thereby discovering more previously unknown triples.Due to the continuous growth of data and knowledge and the limitations of data sources,the visual knowledge within the knowledge graphs is generally of low quality,and some entities suffer from the issue of missing visual modality.Nevertheless,previous studies of MMKGC have primarily focused on how to facilitate modality interaction and fusion while neglecting the problems of low modality quality and modality missing.In this case,mainstream MMKGC models only use pre-trained visual encoders to extract features and transfer the semantic information to the joint embeddings through modal fusion,which inevitably suffers from problems such as error propagation and increased uncertainty.To address these problems,we propose a Multi-modal knowledge graph Completion model based on Super-resolution and Detailed Description Generation(MMCSD).Specifically,we leverage a pre-trained residual network to enhance the resolution and improve the quality of the visual modality.Moreover,we design multi-level visual semantic extraction and entity description generation,thereby further extracting entity semantics from structural triples and visual images.Meanwhile,we train a variational multi-modal auto-encoder and utilize a pre-trained multi-modal language model to complement the missing visual features.We conducted experiments on FB15K-237 and DB13K,and the results showed that MMCSD can effectively perform MMKGC and achieve state-of-the-art performance.展开更多
Utilizing graph neural networks for knowledge embedding to accomplish the task of knowledge graph completion(KGC)has become an important research area in knowledge graph completion.However,the number of nodes in the k...Utilizing graph neural networks for knowledge embedding to accomplish the task of knowledge graph completion(KGC)has become an important research area in knowledge graph completion.However,the number of nodes in the knowledge graph increases exponentially with the depth of the tree,whereas the distances of nodes in Euclidean space are second-order polynomial distances,whereby knowledge embedding using graph neural networks in Euclidean space will not represent the distances between nodes well.This paper introduces a novel approach called hyperbolic hierarchical graph attention network(H2GAT)to rectify this limitation.Firstly,the paper conducts knowledge representation in the hyperbolic space,effectively mitigating the issue of exponential growth of nodes with tree depth and consequent information loss.Secondly,it introduces a hierarchical graph atten-tion mechanism specifically designed for the hyperbolic space,allowing for enhanced capture of the network structure inherent in the knowledge graph.Finally,the efficacy of the proposed H2GAT model is evaluated on benchmark datasets,namely WN18RR and FB15K-237,thereby validating its effectiveness.The H2GAT model achieved 0.445,0.515,and 0.586 in the Hits@1,Hits@3 and Hits@10 metrics respectively on the WN18RR dataset and 0.243,0.367 and 0.518 on the FB15K-237 dataset.By incorporating hyperbolic space embedding and hierarchical graph attention,the H2GAT model successfully addresses the limitations of existing hyperbolic knowledge embedding models,exhibiting its competence in knowledge graph completion tasks.展开更多
To solve the problem of missing many valid triples in knowledge graphs(KGs),a novel model based on a convolutional neural network(CNN)called ConvKG is proposed,which employs a joint learning strategy for knowledge gra...To solve the problem of missing many valid triples in knowledge graphs(KGs),a novel model based on a convolutional neural network(CNN)called ConvKG is proposed,which employs a joint learning strategy for knowledge graph completion(KGC).Related research work has shown the superiority of convolutional neural networks(CNNs)in extracting semantic features of triple embeddings.However,these researches use only one single-shaped filter and fail to extract semantic features of different granularity.To solve this problem,ConvKG exploits multi-shaped filters to co-convolute on the triple embeddings,joint learning semantic features of different granularity.Different shaped filters cover different sizes on the triple embeddings and capture pairwise interactions of different granularity among triple elements.Experimental results confirm the strength of joint learning,and compared with state-of-the-art CNN-based KGC models,ConvKG achieves the better mean rank(MR)and Hits@10 metrics on dataset WN18 RR,and the better MR on dataset FB15k-237.展开更多
Knowledge graph completion(KGC)aims to fill in missing entities and relations within knowledge graphs(KGs)to address their incompleteness.Most existing KGC models suffer from knowledge coverage as they are designed to...Knowledge graph completion(KGC)aims to fill in missing entities and relations within knowledge graphs(KGs)to address their incompleteness.Most existing KGC models suffer from knowledge coverage as they are designed to operate within a single KG.In contrast,Multilingual KGC(MKGC)leverages seed pairs from different language KGs to facilitate knowledge transfer and enhance the completion of the target KG.Previous studies on MKGC based on graph neural networks(GNNs)have primarily focused on using relationaware GNNs to capture the combined features of neighboring entities and relations.However,these studies still have some shortcomings,particularly in the context of MKGCs.First,each language’s specific semantics,structures,and expressions contribute to the increased heterogeneity of the KG.Therefore,the completion of MKGCs necessitates a thorough consideration of the heterogeneity of the KG and the effective integration of its heterogeneous features.Second,MKGCs typically have a large graph scale due to the need to store and manage information from multiple languages.However,current relation-aware GNNs often inherit complex GNN operations,resulting in unnecessary complexity.Therefore,it is necessary to simplify GNN operations.To address these limitations,we propose a Simplified Multi-view Graph Neural Network(SMGNN)for MKGC.SM-GNN incorporates two simplified multiview GNNs as components.One GNN is utilized for learning multi-view graph features to complete the KG.The other generates new alignment pairs,facilitating knowledge transfer between different views of the KG.We simplify the two multiview GNNs by retaining feature propagation while discarding linear transformation and nonlinear activation to reduce unnecessary complexity and effectively leverage graph contextual information.Extensive experiments demonstrate that our proposed model outperforms competing baselines.The code and dataset are available at the website of github.com/dbbice/SM-GNN.展开更多
Sparse Knowledge Graph(KG)scenarios pose a challenge for previous Knowledge Graph Completion(KGC)methods,that is,the completion performance decreases rapidly with the increase of graph sparsity.This problem is also ex...Sparse Knowledge Graph(KG)scenarios pose a challenge for previous Knowledge Graph Completion(KGC)methods,that is,the completion performance decreases rapidly with the increase of graph sparsity.This problem is also exacerbated because of the widespread existence of sparse KGs in practical applications.To alleviate this challenge,we present a novel framework,LR-GCN,that is able to automatically capture valuable long-range dependency among entities to supplement insufficient structure features and distill logical reasoning knowledge for sparse KGC.The proposed approach comprises two main components:a GNN-based predictor and a reasoning path distiller.The reasoning path distiller explores high-order graph structures such as reasoning paths and encodes them as rich-semantic edges,explicitly compositing long-range dependencies into the predictor.This step also plays an essential role in densifying KGs,effectively alleviating the sparse issue.Furthermore,the path distiller further distills logical reasoning knowledge from these mined reasoning paths into the predictor.These two components are jointly optimized using a well-designed variational EM algorithm.Extensive experiments and analyses on four sparse benchmarks demonstrate the effectiveness of our proposed method.展开更多
The objective of knowledge graph completion is to comprehend the structure and inherent relationships of domain knowledge,thereby providing a valuable foundation for knowledge reasoning and analysis.However,existing m...The objective of knowledge graph completion is to comprehend the structure and inherent relationships of domain knowledge,thereby providing a valuable foundation for knowledge reasoning and analysis.However,existing methods for knowledge graph completion face challenges.For instance,rule-based completion methods exhibit high accuracy and interpretability,but encounter difficulties when handling large knowledge graphs.In contrast,embedding-based completion methods demonstrate strong scalability and efficiency,but also have limited utilisation of domain knowledge.In response to the aforementioned issues,we propose a method of pre-training and inference for knowledge graph completion based on integrated rules.The approach combines rule mining and reasoning to generate precise candidate facts.Subsequently,a pre-trained language model is fine-tuned and probabilistic structural loss is incorporated to embed the knowledge graph.This enables the language model to capture more deep semantic information while the loss function reconstructs the structure of the knowledge graph.This enables the language model to capture more deep semantic information while the loss function reconstructs the structure of the knowledge graph.Extensive tests using various publicly accessible datasets have indicated that the suggested model performs better than current techniques in tackling knowledge graph completion problems.展开更多
In the link prediction task of knowledge graph completion,Graph Neural Network(GNN)-based knowledge graph completion models have been shown by previous studies to produce large improvements in prediction results.Howev...In the link prediction task of knowledge graph completion,Graph Neural Network(GNN)-based knowledge graph completion models have been shown by previous studies to produce large improvements in prediction results.However,many of the previous efforts were limited to aggregating the information given by neighboring nodes and did not take advantage of the information provided by the edges represented by relations.To address the problem,Coupling Relation Strength with Graph Convolutional Networks(RS-GCN)is proposed,which is a model with an encoder-decoder framework to realize the embedding of entities and relations in the vector space.On the encoder side,RS-GCN captures graph structure and neighborhood information while aggregating the information given by neighboring nodes.On the decoder side,RotatE is utilized to model and infer various relational patterns.The models are evaluated on standard FB15k,WN18,FB15k-237 and WN18RR datasets,and the experiments show that RS-GCN achieves better results than the current state-of-the-art classical models on the above knowledge graph datasets.展开更多
Knowledge Graph Completion(KGC)aims to predict missing links in a knowledge graph.A popular model for this task is the Graph Neural Network(GNN),which leverages structural information from neighboring nodes.However,cu...Knowledge Graph Completion(KGC)aims to predict missing links in a knowledge graph.A popular model for this task is the Graph Neural Network(GNN),which leverages structural information from neighboring nodes.However,current GNN-based methods treat all neighbors equally,overlooking the importance of entity neighbors and handling complex relationships effectively.To address these challenges,we introduce the Hierarchical Entity Neighbor Multi-Relational Fusion Network(HENF)for KGC.HENF offers fine-grained adaptability to various multi-relational scenarios.It constructs relationship subgraphs based on one-hop paths between entities,aggregating information around entities using dynamic attention mechanisms.Furthermore,it employs Adjacent Relation Fusion(ARF)attention to combine rich entity information from different relational graphs.This approach allows our model to emphasize diverse semantic information types under various relations,selectively gather informative features,and assign appropriate weights.Extensive experiments demonstrate that HENF significantly enhances KGC performance,especially on datasets with many-to-many relationships.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
In the context of big data, many large-scale knowledge graphs have emerged to effectively organize the explosive growth of web data on the Internet. To select suitable knowledge graphs for use from many knowledge grap...In the context of big data, many large-scale knowledge graphs have emerged to effectively organize the explosive growth of web data on the Internet. To select suitable knowledge graphs for use from many knowledge graphs, quality assessment is particularly important. As an important thing of quality assessment, completeness assessment generally refers to the ratio of the current data volume to the total data volume.When evaluating the completeness of a knowledge graph, it is often necessary to refine the completeness dimension by setting different completeness metrics to produce more complete and understandable evaluation results for the knowledge graph.However, lack of awareness of requirements is the most problematic quality issue. In the actual evaluation process, the existing completeness metrics need to consider the actual application. Therefore, to accurately recommend suitable knowledge graphs to many users, it is particularly important to develop relevant measurement metrics and formulate measurement schemes for completeness. In this paper, we will first clarify the concept of completeness, establish each metric of completeness, and finally design a measurement proposal for the completeness of knowledge graphs.展开更多
Knowledge graph(KG) link prediction aims to address the problem of missing multiple valid triples in KGs. Existing approaches either struggle to efficiently model the message passing process of multi-hop paths or lack...Knowledge graph(KG) link prediction aims to address the problem of missing multiple valid triples in KGs. Existing approaches either struggle to efficiently model the message passing process of multi-hop paths or lack transparency of model prediction principles. In this paper,a new graph convolutional network path semantic-aware graph convolution network(PSGCN) is proposed to achieve modeling the semantic information of multi-hop paths. PSGCN first uses a random walk strategy to obtain all-hop paths in KGs,then captures the semantics of the paths by Word2Sec and long shortterm memory(LSTM) models,and finally converts them into a potential representation for the graph convolution network(GCN) messaging process. PSGCN combines path-based inference methods and graph neural networks to achieve better interpretability and scalability. In addition,to ensure the robustness of the model,the value of the path thresholdKis experimented on the FB15K-237 and WN18RR datasets,and the final results prove the effectiveness of the model.展开更多
Knowledge representation learning(KRL)aims to encode entities and relationships in various knowledge graphs into low-dimensional continuous vectors.It is popularly used in knowledge graph completion(or link prediction...Knowledge representation learning(KRL)aims to encode entities and relationships in various knowledge graphs into low-dimensional continuous vectors.It is popularly used in knowledge graph completion(or link prediction)tasks.Translation-based knowledge representation learning methods perform well in knowledge graph completion(KGC).However,the translation principles adopted by these methods are too strict and cannot model complex entities and relationships(i.e.,N-1,1-N,and N-N)well.Besides,these traditional translation principles are primarily used in static knowledge graphs and overlook the temporal properties of triplet facts.Therefore,we propose a temporal knowledge graph embedding model based on variable translation(TKGE-VT).The model proposes a new variable translation principle,which enables flexible transformation between entities and relationship embedding.Meanwhile,this paper considers the temporal properties of both entities and relationships and applies the proposed principle of variable translation to temporal knowledge graphs.We conduct link prediction and triplet classification experiments on four benchmark datasets:WN11,WN18,FB13,and FB15K.Our model outperforms baseline models on multiple evaluation metrics according to the experimental results.展开更多
Knowledge graphs are involved in more and more applications to further improve intelligence.Owing to the inherent incompleteness of knowledge graphs resulted from data updating and missing,a number of knowledge graph ...Knowledge graphs are involved in more and more applications to further improve intelligence.Owing to the inherent incompleteness of knowledge graphs resulted from data updating and missing,a number of knowledge graph completion models are proposed in succession.To obtain better performance,many methods are of high complexity,making it time-consuming for training and inference.This paper proposes a simple but e®ective model using only shallow neural networks,which combines enhanced feature interaction and multi-subspace information integration.In the enhanced feature interaction module,entity and relation embeddings are almost peer-to-peer interacted via multi-channel 2D convolution.In the multi-subspace information integration module,entity and relation embeddings are projected to multiple subspaces to extract multi-view information to further boost performance.Extensive experiments on widely used datasets show that the proposed model outperforms a series of strong baselines.And ablation studies demonstrate the e®ectiveness of each submodule in the model.展开更多
Currently,most existing inductive relation prediction approaches are based on subgraph structures,with subgraph features extracted using graph neural networks to predict relations.However,subgraphs may contain disconn...Currently,most existing inductive relation prediction approaches are based on subgraph structures,with subgraph features extracted using graph neural networks to predict relations.However,subgraphs may contain disconnected regions,which usually represent different semantic ranges.Because not all semantic information about the regions is helpful in relation prediction,we propose a relation prediction model based on a disentangled subgraph structure and implement a feature updating approach based on relevant semantic aggregation.To indirectly achieve the disentangled subgraph structure from a semantic perspective,the mapping of entity features into different semantic spaces and the aggregation of related semantics on each semantic space are updated.The disentangled model can focus on features having higher semantic relevance in the prediction,thus addressing a problem with existing approaches,which ignore the semantic differences in different subgraph structures.Furthermore,using a gated recurrent neural network,this model enhances the features of entities by sorting them by distance and extracting the path information in the subgraphs.Experimentally,it is shown that when there are numerous disconnected regions in the subgraph,our model outperforms existing mainstream models in terms of both Area Under the Curve-Precision-Recall(AUC-PR)and Hits@10.Experiments prove that semantic differences in the knowledge graph can be effectively distinguished and verify the effectiveness of this method.展开更多
The Knowledge Graph(KGs)have profoundly impacted many researchfields.However,there is a problem of low data integrity in KGs.The binary-relational knowledge graph is more common in KGs but is limited by less informatio...The Knowledge Graph(KGs)have profoundly impacted many researchfields.However,there is a problem of low data integrity in KGs.The binary-relational knowledge graph is more common in KGs but is limited by less information.It often has less content to use when predicting missing entities(relations).The hyper-relational knowledge graph is another form of KGs,which introduces much additional information(qualifiers)based on the main triple.The hyper-relational knowledge graph can effectively improve the accuracy of pre-dicting missing entities(relations).The existing hyper-relational link prediction methods only consider the overall perspective when dealing with qualifiers and calculate the score function by combining the qualifiers with the main triple.How-ever,these methods overlook the inherent characteristics of entities and relations.This paper proposes a novel Local and Global Hyper-relation Aggregation Embed-ding for Link Prediction(LGHAE).LGHAE can capture the semantic features of hyper-relational data from local and global perspectives.To fully utilize local and global features,Hyper-InteractE,as a new decoder,is designed to predict missing entities to fully utilize local and global features.We validated the feasibility of LGHAE by comparing it with state-of-the-art models on public datasets.展开更多
基金funded by Research Project,grant number BHQ090003000X03。
文摘Multi-modal knowledge graph completion(MMKGC)aims to complete missing entities or relations in multi-modal knowledge graphs,thereby discovering more previously unknown triples.Due to the continuous growth of data and knowledge and the limitations of data sources,the visual knowledge within the knowledge graphs is generally of low quality,and some entities suffer from the issue of missing visual modality.Nevertheless,previous studies of MMKGC have primarily focused on how to facilitate modality interaction and fusion while neglecting the problems of low modality quality and modality missing.In this case,mainstream MMKGC models only use pre-trained visual encoders to extract features and transfer the semantic information to the joint embeddings through modal fusion,which inevitably suffers from problems such as error propagation and increased uncertainty.To address these problems,we propose a Multi-modal knowledge graph Completion model based on Super-resolution and Detailed Description Generation(MMCSD).Specifically,we leverage a pre-trained residual network to enhance the resolution and improve the quality of the visual modality.Moreover,we design multi-level visual semantic extraction and entity description generation,thereby further extracting entity semantics from structural triples and visual images.Meanwhile,we train a variational multi-modal auto-encoder and utilize a pre-trained multi-modal language model to complement the missing visual features.We conducted experiments on FB15K-237 and DB13K,and the results showed that MMCSD can effectively perform MMKGC and achieve state-of-the-art performance.
基金the Beijing Municipal Science and Technology Program(No.Z231100001323004).
文摘Utilizing graph neural networks for knowledge embedding to accomplish the task of knowledge graph completion(KGC)has become an important research area in knowledge graph completion.However,the number of nodes in the knowledge graph increases exponentially with the depth of the tree,whereas the distances of nodes in Euclidean space are second-order polynomial distances,whereby knowledge embedding using graph neural networks in Euclidean space will not represent the distances between nodes well.This paper introduces a novel approach called hyperbolic hierarchical graph attention network(H2GAT)to rectify this limitation.Firstly,the paper conducts knowledge representation in the hyperbolic space,effectively mitigating the issue of exponential growth of nodes with tree depth and consequent information loss.Secondly,it introduces a hierarchical graph atten-tion mechanism specifically designed for the hyperbolic space,allowing for enhanced capture of the network structure inherent in the knowledge graph.Finally,the efficacy of the proposed H2GAT model is evaluated on benchmark datasets,namely WN18RR and FB15K-237,thereby validating its effectiveness.The H2GAT model achieved 0.445,0.515,and 0.586 in the Hits@1,Hits@3 and Hits@10 metrics respectively on the WN18RR dataset and 0.243,0.367 and 0.518 on the FB15K-237 dataset.By incorporating hyperbolic space embedding and hierarchical graph attention,the H2GAT model successfully addresses the limitations of existing hyperbolic knowledge embedding models,exhibiting its competence in knowledge graph completion tasks.
基金Supported by the National Natural Science Foundation of China(No.61876144)。
文摘To solve the problem of missing many valid triples in knowledge graphs(KGs),a novel model based on a convolutional neural network(CNN)called ConvKG is proposed,which employs a joint learning strategy for knowledge graph completion(KGC).Related research work has shown the superiority of convolutional neural networks(CNNs)in extracting semantic features of triple embeddings.However,these researches use only one single-shaped filter and fail to extract semantic features of different granularity.To solve this problem,ConvKG exploits multi-shaped filters to co-convolute on the triple embeddings,joint learning semantic features of different granularity.Different shaped filters cover different sizes on the triple embeddings and capture pairwise interactions of different granularity among triple elements.Experimental results confirm the strength of joint learning,and compared with state-of-the-art CNN-based KGC models,ConvKG achieves the better mean rank(MR)and Hits@10 metrics on dataset WN18 RR,and the better MR on dataset FB15k-237.
基金supported by the National Natural Science Foundation of China(Grant Nos.62120106008,61976077,61806065,62076085,and 91746209)the Fundamental Research Funds for the Central Universities(JZ2022HGTB0239).
文摘Knowledge graph completion(KGC)aims to fill in missing entities and relations within knowledge graphs(KGs)to address their incompleteness.Most existing KGC models suffer from knowledge coverage as they are designed to operate within a single KG.In contrast,Multilingual KGC(MKGC)leverages seed pairs from different language KGs to facilitate knowledge transfer and enhance the completion of the target KG.Previous studies on MKGC based on graph neural networks(GNNs)have primarily focused on using relationaware GNNs to capture the combined features of neighboring entities and relations.However,these studies still have some shortcomings,particularly in the context of MKGCs.First,each language’s specific semantics,structures,and expressions contribute to the increased heterogeneity of the KG.Therefore,the completion of MKGCs necessitates a thorough consideration of the heterogeneity of the KG and the effective integration of its heterogeneous features.Second,MKGCs typically have a large graph scale due to the need to store and manage information from multiple languages.However,current relation-aware GNNs often inherit complex GNN operations,resulting in unnecessary complexity.Therefore,it is necessary to simplify GNN operations.To address these limitations,we propose a Simplified Multi-view Graph Neural Network(SMGNN)for MKGC.SM-GNN incorporates two simplified multiview GNNs as components.One GNN is utilized for learning multi-view graph features to complete the KG.The other generates new alignment pairs,facilitating knowledge transfer between different views of the KG.We simplify the two multiview GNNs by retaining feature propagation while discarding linear transformation and nonlinear activation to reduce unnecessary complexity and effectively leverage graph contextual information.Extensive experiments demonstrate that our proposed model outperforms competing baselines.The code and dataset are available at the website of github.com/dbbice/SM-GNN.
基金supported by the National Key R&D Program of China(2022YFF0903301)the National Natural Science Foundation of China(Grant Nos.U22B2059,61976073,62276083)+1 种基金the Shenzhen Foundational Research Funding(JCYJ20200109113441941)the Major Key Project of PCL(PCL2021A06).
文摘Sparse Knowledge Graph(KG)scenarios pose a challenge for previous Knowledge Graph Completion(KGC)methods,that is,the completion performance decreases rapidly with the increase of graph sparsity.This problem is also exacerbated because of the widespread existence of sparse KGs in practical applications.To alleviate this challenge,we present a novel framework,LR-GCN,that is able to automatically capture valuable long-range dependency among entities to supplement insufficient structure features and distill logical reasoning knowledge for sparse KGC.The proposed approach comprises two main components:a GNN-based predictor and a reasoning path distiller.The reasoning path distiller explores high-order graph structures such as reasoning paths and encodes them as rich-semantic edges,explicitly compositing long-range dependencies into the predictor.This step also plays an essential role in densifying KGs,effectively alleviating the sparse issue.Furthermore,the path distiller further distills logical reasoning knowledge from these mined reasoning paths into the predictor.These two components are jointly optimized using a well-designed variational EM algorithm.Extensive experiments and analyses on four sparse benchmarks demonstrate the effectiveness of our proposed method.
基金supported by the Sichuan Science and Technology Program(No.2024YFHZ0024)the National Natural Science Foundation of China(No.62276215)。
文摘The objective of knowledge graph completion is to comprehend the structure and inherent relationships of domain knowledge,thereby providing a valuable foundation for knowledge reasoning and analysis.However,existing methods for knowledge graph completion face challenges.For instance,rule-based completion methods exhibit high accuracy and interpretability,but encounter difficulties when handling large knowledge graphs.In contrast,embedding-based completion methods demonstrate strong scalability and efficiency,but also have limited utilisation of domain knowledge.In response to the aforementioned issues,we propose a method of pre-training and inference for knowledge graph completion based on integrated rules.The approach combines rule mining and reasoning to generate precise candidate facts.Subsequently,a pre-trained language model is fine-tuned and probabilistic structural loss is incorporated to embed the knowledge graph.This enables the language model to capture more deep semantic information while the loss function reconstructs the structure of the knowledge graph.This enables the language model to capture more deep semantic information while the loss function reconstructs the structure of the knowledge graph.Extensive tests using various publicly accessible datasets have indicated that the suggested model performs better than current techniques in tackling knowledge graph completion problems.
文摘In the link prediction task of knowledge graph completion,Graph Neural Network(GNN)-based knowledge graph completion models have been shown by previous studies to produce large improvements in prediction results.However,many of the previous efforts were limited to aggregating the information given by neighboring nodes and did not take advantage of the information provided by the edges represented by relations.To address the problem,Coupling Relation Strength with Graph Convolutional Networks(RS-GCN)is proposed,which is a model with an encoder-decoder framework to realize the embedding of entities and relations in the vector space.On the encoder side,RS-GCN captures graph structure and neighborhood information while aggregating the information given by neighboring nodes.On the decoder side,RotatE is utilized to model and infer various relational patterns.The models are evaluated on standard FB15k,WN18,FB15k-237 and WN18RR datasets,and the experiments show that RS-GCN achieves better results than the current state-of-the-art classical models on the above knowledge graph datasets.
文摘Knowledge Graph Completion(KGC)aims to predict missing links in a knowledge graph.A popular model for this task is the Graph Neural Network(GNN),which leverages structural information from neighboring nodes.However,current GNN-based methods treat all neighbors equally,overlooking the importance of entity neighbors and handling complex relationships effectively.To address these challenges,we introduce the Hierarchical Entity Neighbor Multi-Relational Fusion Network(HENF)for KGC.HENF offers fine-grained adaptability to various multi-relational scenarios.It constructs relationship subgraphs based on one-hop paths between entities,aggregating information around entities using dynamic attention mechanisms.Furthermore,it employs Adjacent Relation Fusion(ARF)attention to combine rich entity information from different relational graphs.This approach allows our model to emphasize diverse semantic information types under various relations,selectively gather informative features,and assign appropriate weights.Extensive experiments demonstrate that HENF significantly enhances KGC performance,especially on datasets with many-to-many relationships.
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supported by the National Key Laboratory for Comp lex Systems Simulation Foundation (6142006190301)。
文摘In the context of big data, many large-scale knowledge graphs have emerged to effectively organize the explosive growth of web data on the Internet. To select suitable knowledge graphs for use from many knowledge graphs, quality assessment is particularly important. As an important thing of quality assessment, completeness assessment generally refers to the ratio of the current data volume to the total data volume.When evaluating the completeness of a knowledge graph, it is often necessary to refine the completeness dimension by setting different completeness metrics to produce more complete and understandable evaluation results for the knowledge graph.However, lack of awareness of requirements is the most problematic quality issue. In the actual evaluation process, the existing completeness metrics need to consider the actual application. Therefore, to accurately recommend suitable knowledge graphs to many users, it is particularly important to develop relevant measurement metrics and formulate measurement schemes for completeness. In this paper, we will first clarify the concept of completeness, establish each metric of completeness, and finally design a measurement proposal for the completeness of knowledge graphs.
基金Supported by the National Natural Science Foundation of China(No.61876144).
文摘Knowledge graph(KG) link prediction aims to address the problem of missing multiple valid triples in KGs. Existing approaches either struggle to efficiently model the message passing process of multi-hop paths or lack transparency of model prediction principles. In this paper,a new graph convolutional network path semantic-aware graph convolution network(PSGCN) is proposed to achieve modeling the semantic information of multi-hop paths. PSGCN first uses a random walk strategy to obtain all-hop paths in KGs,then captures the semantics of the paths by Word2Sec and long shortterm memory(LSTM) models,and finally converts them into a potential representation for the graph convolution network(GCN) messaging process. PSGCN combines path-based inference methods and graph neural networks to achieve better interpretability and scalability. In addition,to ensure the robustness of the model,the value of the path thresholdKis experimented on the FB15K-237 and WN18RR datasets,and the final results prove the effectiveness of the model.
基金supported partly by National Natural Science Foundation of China(Nos.62372119 and 62166003)the Project of Guangxi Science and Technology(Nos.GuiKeAB23026040 and GuiKeAD20159041)+3 种基金the Innovation Project of Guangxi Graduate Education(No.YCSW2023188)Key Lab of Education Blockchain and Intelligent Technology,Ministry of Education,Guangxi Normal University,Guilin,China,Intelligent Processing and the Research Fund of Guangxi Key Lab of Multi-source Information Mining&Security(Nos.20-A-01-01 and MIMS21-M01)Open Research Fund of Guangxi Key Lab of Humanmachine Interaction and Intelligent Decision(No.GXHIID2206)the Guangxi Collaborative Innovation Center of Multi-Source Information Integration and the Guangxi“Bagui”Teams for Innovation and Research,China.
文摘Knowledge representation learning(KRL)aims to encode entities and relationships in various knowledge graphs into low-dimensional continuous vectors.It is popularly used in knowledge graph completion(or link prediction)tasks.Translation-based knowledge representation learning methods perform well in knowledge graph completion(KGC).However,the translation principles adopted by these methods are too strict and cannot model complex entities and relationships(i.e.,N-1,1-N,and N-N)well.Besides,these traditional translation principles are primarily used in static knowledge graphs and overlook the temporal properties of triplet facts.Therefore,we propose a temporal knowledge graph embedding model based on variable translation(TKGE-VT).The model proposes a new variable translation principle,which enables flexible transformation between entities and relationship embedding.Meanwhile,this paper considers the temporal properties of both entities and relationships and applies the proposed principle of variable translation to temporal knowledge graphs.We conduct link prediction and triplet classification experiments on four benchmark datasets:WN11,WN18,FB13,and FB15K.Our model outperforms baseline models on multiple evaluation metrics according to the experimental results.
基金the National Natural Science Foundation of China under Grant No.61991412the Program for HUST Academic Frontier Youth Team under Grant No.2018QYTD07.
文摘Knowledge graphs are involved in more and more applications to further improve intelligence.Owing to the inherent incompleteness of knowledge graphs resulted from data updating and missing,a number of knowledge graph completion models are proposed in succession.To obtain better performance,many methods are of high complexity,making it time-consuming for training and inference.This paper proposes a simple but e®ective model using only shallow neural networks,which combines enhanced feature interaction and multi-subspace information integration.In the enhanced feature interaction module,entity and relation embeddings are almost peer-to-peer interacted via multi-channel 2D convolution.In the multi-subspace information integration module,entity and relation embeddings are projected to multiple subspaces to extract multi-view information to further boost performance.Extensive experiments on widely used datasets show that the proposed model outperforms a series of strong baselines.And ablation studies demonstrate the e®ectiveness of each submodule in the model.
基金supported by the National Natural Science Foundation of China(No.U19A2059)the 2022 Research Foundation of Chengdu Textile College(No.X22032161).
文摘Currently,most existing inductive relation prediction approaches are based on subgraph structures,with subgraph features extracted using graph neural networks to predict relations.However,subgraphs may contain disconnected regions,which usually represent different semantic ranges.Because not all semantic information about the regions is helpful in relation prediction,we propose a relation prediction model based on a disentangled subgraph structure and implement a feature updating approach based on relevant semantic aggregation.To indirectly achieve the disentangled subgraph structure from a semantic perspective,the mapping of entity features into different semantic spaces and the aggregation of related semantics on each semantic space are updated.The disentangled model can focus on features having higher semantic relevance in the prediction,thus addressing a problem with existing approaches,which ignore the semantic differences in different subgraph structures.Furthermore,using a gated recurrent neural network,this model enhances the features of entities by sorting them by distance and extracting the path information in the subgraphs.Experimentally,it is shown that when there are numerous disconnected regions in the subgraph,our model outperforms existing mainstream models in terms of both Area Under the Curve-Precision-Recall(AUC-PR)and Hits@10.Experiments prove that semantic differences in the knowledge graph can be effectively distinguished and verify the effectiveness of this method.
文摘The Knowledge Graph(KGs)have profoundly impacted many researchfields.However,there is a problem of low data integrity in KGs.The binary-relational knowledge graph is more common in KGs but is limited by less information.It often has less content to use when predicting missing entities(relations).The hyper-relational knowledge graph is another form of KGs,which introduces much additional information(qualifiers)based on the main triple.The hyper-relational knowledge graph can effectively improve the accuracy of pre-dicting missing entities(relations).The existing hyper-relational link prediction methods only consider the overall perspective when dealing with qualifiers and calculate the score function by combining the qualifiers with the main triple.How-ever,these methods overlook the inherent characteristics of entities and relations.This paper proposes a novel Local and Global Hyper-relation Aggregation Embed-ding for Link Prediction(LGHAE).LGHAE can capture the semantic features of hyper-relational data from local and global perspectives.To fully utilize local and global features,Hyper-InteractE,as a new decoder,is designed to predict missing entities to fully utilize local and global features.We validated the feasibility of LGHAE by comparing it with state-of-the-art models on public datasets.
