Entities In Knowledge Graphs Research Articles

Atom: An Efficient Query Serving System for Embedding-based Knowledge Graph Reasoning with Operator-level Batching

Knowledge graph reasoning (KGR) answers logical queries over a knowledge graph (KG), and embedding-based KGR (EKGR) becomes popular recently, which embeds both queries and KG entities such that the vector embeddings of a query and its answer entities are similar. Compared with traditional KGR methods based on subgraph matching, EKGR produces fewer intermediate results and is more robust to missing and noisy information in the KG. However, existing systems are inefficient for serving online EKGR queries because they can only batch queries of the same type for execution (i.e., query-level batching ) and hence have limited batching opportunities due to the heterogeneity of queries. To serve EKGR queries efficiently, we propose the Atom system with operator-level batching, which decomposes queries into operators and batches operators of the same type from different queries for execution. The insight is that the types of operators are far fewer than the types of queries, and thus different queries typically share common operators, yielding more batching opportunities. To schedule the operators, Atom adopts a hybrid policy, which improves system throughput and avoids starving rare operators. For efficiency, Atom incorporates system optimizations including two-level pipeline, opportunistic submission, pre-allocated memory buffer, and tailored GPU kernels. Experiment results show that compared with existing systems, Atom can improve query throughput by over 20x and reduce query latency by over 5x. Micro experiments suggest that the designs and optimizations are effective in improving system performance.

A semi-supervised framework fusing multiple information for knowledge graph entity alignment

Entity alignment (EA) is a fundamental task for cross linguistic knowledge graphs (KGs) understanding and interaction, which is committed to matching entities from different graphs based on their inherent semantics. Methods based on graph neural networks (GNNs) dominate the EA task, however, the majority of them ignore the higher-order information among entities in the KGs. Meanwhile, as important auxiliary information, the relational semantics, string information of entity names and attribute information of entities are insufficiently exploited during the inference phase. In addition, labeled alignment data is universally insufficient across various datasets, which limits the performance of the model. In this paper, we propose a Semi-supervised EA framework that Comprehensively considers both Structural and Attribute information within KGs (SCSA) to address these problems above. Specifically, our approach first leverages hypergraph neural networks (HGNN) to aggregate relational semantic information and graph convolutional networks (GCNs) with a highway filtering strategy to acquire the embedding representation of entities precisely. Then, we propose a bidirectional filtering technique with a combination of entity, attribute and string values to create pseudo-labeled data and lead the model for iteratively training. We implement our proposed framework on several publicly recognized cross-lingual datasets. The experimental results indicate that our framework outperforms almost all state-of-the-art (SOTA) methods.

A survey: knowledge graph entity alignment research based on graph embedding

Entity alignment (EA) aims to automatically match entities in different knowledge graphs, which is beneficial to the development of knowledge-driven applications. Representation learning has powerful feature capture capability and it is widely used in the field of natural language processing. Compared with traditional EA methods, EA methods based on representation learning have better performance and efficiency. Hence, we summarize and analyze the representative EA approaches based on representation learning in this paper. We present the problem description and data preprocessing for EA and other related fundamental knowledge. We propose a new EA framework for the latest models, which includes information aggregation module, entity alignment module, and post-alignment module. Based on these three modules, the various technologies are described in detail. In the experimental part, we first explore the effect of EA direction on model performance. Then, we classify the models into different categories in terms of alignment inference strategy, noise filtering strategy, and whether additional information is utilized. To ensure fairness, we perform the comparative analysis of the performance of the models within the categories separately on different datasets. We investigate both unimodal and multimodal EA. Finally, we present future research perspectives based on the shortcomings of existing EA methods.

Multi-stream graph attention network for recommendation with knowledge graph

In recent years, the powerful modeling ability of Graph Neural Networks (GNNs) has led to their widespread use in knowledge-aware recommender systems. However, existing GNN-based methods for information propagation among entities in knowledge graphs (KGs) may not efficiently filter out less informative entities. To address this challenge and improve the encoding of high-order structure information among many entities, we propose an end-to-end neural network-based method called Multi-stream Graph Attention Network (MSGAT). MSGAT explicitly discriminates the importance of entities from four critical perspectives and recursively propagates neighbor embeddings to refine the target node. Specifically, we use an attention mechanism from the user's perspective to distill the domain nodes' information of the predicted item in the KG, enhance the user's information on items, and generate the feature representation of the predicted item. We also propose a multi-stream attention mechanism to aggregate user history click item's neighborhood entity information in the KG and generate the user's feature representation. We conduct extensive experiments on three real datasets for movies, music, and books, and the empirical results demonstrate that MSGAT outperforms current state-of-the-art baselines.

Triplet-aware graph neural networks for factorized multi-modal knowledge graph entity alignment

Multi-Modal Entity Alignment (MMEA), aiming to discover matching entity pairs on two multi-modal knowledge graphs (MMKGs), is an essential task in knowledge graph fusion. Through mining feature information of MMKGs, entities are aligned to tackle the issue that an MMKG is incapable of effective integration. The recent attempt at neighbors and attribute fusion mainly focuses on aggregating multi-modal attributes, neglecting the structure effect with multi-modal attributes for entity alignment. This paper proposes an innovative approach, namely TriFac, to exploit embedding refinement for factorizing the original multi-modal knowledge graphs through a two-stage MMKG factorization. Notably, we propose triplet-aware graph neural networks to aggregate multi-relational features. We propose multi-modal fusion for aggregating multiple features and design three novel metrics to measure knowledge graph factorization performance on the unified factorized latent space. Empirical results indicate the effectiveness of TriFac, surpassing previous state-of-the-art models on two MMEA datasets and a power system dataset.

Entity type inference based on path walking and inter-types relationships

As a crucial task for knowledge graphs (KGs), knowledge graph entity type inference (KGET) has garnered increasing attention in recent years. However, recent methods overlook the long-distance information pertaining to entities and the inter-types relationships. The neglect of long-distance information results in the omission of crucial entity relationships and neighbors, consequently leading to the loss of path information associated with missing types. To address this, a path-walking strategy is utilized to identify two-hop triplet paths of the crucial entity for encoding long-distance entity information. Moreover, the absence of inter-types relationships can lead to the loss of the neighborhood information of types, such as co-occurrence information. To ensure a comprehensive understanding of inter-types relationships, we consider interactions not only with the types of single entity but also with different types of entities. Finally, in order to comprehensively represent entities for missing types, considering both the dimensions of path information and neighborhood information, we propose an entity type inference model based on path walking and inter-types relationships, denoted as “ET-PT”. This model effectively extracts comprehensive entity information, thereby obtaining the most complete semantic representation of entities. The experimental results on publicly available datasets demonstrate that the proposed method outperforms state-of-the-art approaches.

Entity and Relation Linking for Knowledge Graph Question Answering Using Gradual Searching

Knowledge graph question answering (KGQA) systems have an important role in retrieving data from a knowledge graph (KG). With the system, regular users can access data from a KG without the need to construct a formal SPARQL query. KGQA systems receive a natural language question (NLQ) and translate it into a SPARQL query through three main tasks, namely, entity and relation detection, entity and relation linking, and query construction. However, the translation is not trivial due to lexical gaps and entity ambiguity that may occur during entity or relation linking. This research proposed an approach based on multiclass classification of NLQ whose entity occurrences are detected into categories based on KG relations to address the lexical gap challenge. Next, to solve the entity ambiguity challenge, this research proposed a three-stage searching procedure to determine appropriate KG entities associated with the NLQ entities, given the correspondence between the NLQ and a particular KG relation. This three-stage searching consisted of text-based searching, vector-based searching, and entity and relation pairing. The proposed approach was evaluated on the SimpleQuestions and LC-QuAD 2.0 datasets. The experiments demonstrated that the proposed approach outperformed the state-of-the-art baseline. For the relation linking task, the proposed approach reached 89.87% and 74.83% recall for the SimpleQuestions and LC-QuAD 2.0, respectively. This approach also achieved 91.74% and 61.96% recall on the entity linking tasks for the SimpleQuestions and LC-QuAD 2.0, respectively.

MDSEA: Knowledge Graph Entity Alignment Based on Multimodal Data Supervision

With the development of social media, the internet, and sensing technologies, multimodal data are becoming increasingly common. Integrating these data into knowledge graphs can help models to better understand and utilize these rich sources of information. The basic idea of the existing methods for entity alignment in knowledge graphs is to extract different data features, such as structure, text, attributes, images, etc., and then fuse these different modal features. The entity similarity in different knowledge graphs is calculated based on the fused features. However, the structures, attribute information, image information, text descriptions, etc., of different knowledge graphs often have significant differences. Directly integrating different modal information can easily introduce noise, thus affecting the effectiveness of the entity alignment. To address the above issues, this paper proposes a knowledge graph entity alignment method based on multimodal data supervision. First, Transformer is used to obtain encoded representations of knowledge graph entities. Then, a multimodal supervised method is used for learning the entity representations in the knowledge graph so that the vector representations of the entities contain rich multimodal semantic information, thereby enhancing the generalization ability of the learned entity representations. Finally, the information from different modalities is mapped to a shared low-dimensional subspace, making similar entities closer in the subspace, thus optimizing the entity alignment effect. The experiments on the DBP15K dataset compared with methods such as MTransE, JAPE, EVA, DNCN, etc., all achieve optimal results.

GLSEC: Global and local semantic-enhanced contrastive framework for knowledge graph completion

Knowledge graph completion (KGC) aims to infer missing links between entities in knowledge graphs (KGs). Recently, models based on graph neural networks (GNNs) have gained widespread attention due to their effectiveness in leveraging the topological structure information of entities. Meanwhile, contrastive learning (CL) has been employed in GNNs-based models to provide more supervised signals for better entity representation in a self-supervised manner. However, existing methods overlook the potential global semantic collaboration among entities within the entire KG. And the application of CL in KGC models often adopt random graph augmentation or basic node structure contrast, leading to suboptimal performance. To tackle them, we propose a Global and Local Semantic-Enhanced Contrastive Framework (GLSEC) for KGC. Specifically, we develop a global Attribute-aware encoder to capture the global semantic features of entities based on an entity-entity Attribute Interaction Graph (AIG). Additionally, we design a Light Graph Aggregation Network (Light-GAN) that innovatively updates the global semantic features using the AIG, combining both efficiency and a lightweight design. Furthermore, we introduce a Global-Local cross-view Contrastive Learning (GLCL) method that contrasts embeddings from global and local views, thereby improving contrastive sample quality and boosting the model’s understanding of entities in various contexts. Extensive experiments show that our model outperforms state-of-the-art KGC methods on benchmark datasets FB15k-237 and WN18RR.

KDGene: knowledge graph completion for disease gene prediction using interactional tensor decomposition.

The accurate identification of disease-associated genes is crucial for understanding the molecular mechanisms underlying various diseases. Most current methods focus on constructing biological networks and utilizing machine learning, particularly deep learning, to identify disease genes. However, these methods overlook complex relations among entities in biological knowledge graphs. Such information has been successfully applied in other areas of life science research, demonstrating their effectiveness. Knowledge graph embedding methods can learn the semantic information of different relations within the knowledge graphs. Nonetheless, the performance of existing representation learning techniques, when applied to domain-specific biological data, remains suboptimal. To solve these problems, we construct a biological knowledge graph centered on diseases and genes, and develop an end-to-end knowledge graph completion framework for disease gene prediction using interactional tensor decomposition named KDGene. KDGene incorporates an interaction module that bridges entity and relation embeddings within tensor decomposition, aiming to improve the representation of semantically similar concepts in specific domains and enhance the ability to accurately predict disease genes. Experimental results show that KDGene significantly outperforms state-of-the-art algorithms, whether existing disease gene prediction methods or knowledge graph embedding methods for general domains. Moreover, the comprehensive biological analysis of the predicted results further validates KDGene's capability to accurately identify new candidate genes. This work proposes a scalable knowledge graph completion framework to identify disease candidate genes, from which the results are promising to provide valuable references for further wet experiments. Data and source codes are available at https://github.com/2020MEAI/KDGene.

A Dual-Way Enhanced Framework from Text Matching Point of View for Multimodal Entity Linking

Multimodal Entity Linking (MEL) aims at linking ambiguous mentions with multimodal information to entity in Knowledge Graph (KG) such as Wikipedia, which plays a key role in many applications. However, existing methods suffer from shortcomings, including modality impurity such as noise in raw image and ambiguous textual entity representation, which puts obstacles to MEL. We formulate multimodal entity linking as a neural text matching problem where each multimodal information (text and image) is treated as a query, and the model learns the mapping from each query to the relevant entity from candidate entities. This paper introduces a dual-way enhanced (DWE) framework for MEL: (1) our model refines queries with multimodal data and addresses semantic gaps using cross-modal enhancers between text and image information. Besides, DWE innovatively leverages fine-grained image attributes, including facial characteristic and scene feature, to enhance and refine visual features. (2)By using Wikipedia descriptions, DWE enriches entity semantics and obtains more comprehensive textual representation, which reduces between textual representation and the entities in KG. Extensive experiments on three public benchmarks demonstrate that our method achieves state-of-the-art (SOTA) performance, indicating the superiority of our model. The code is released on https://github.com/season1blue/DWE.

Select and Augment: Enhanced Dense Retrieval Knowledge Graph Augmentation (Abstract Reprint)

Injecting textual information into knowledge graph (KG) entity representations has been a worthwhile expedition in terms of improving performance in KG oriented tasks within the NLP community. External knowledge often adopted to enhance KG embeddings ranges from semantically rich lexical dependency parsed features to a set of relevant key words to entire text descriptions supplied from an external corpus such as wikipedia and many more. Despite the gains this innovation (Text-enhanced KG embeddings) has made, the proposal in this work suggests that it can be improved even further. Instead of using a single text description (which would not sufficiently represent an entity because of the inherent lexical ambiguity of text), we propose a multi-task framework that jointly selects a set of text descriptions relevant to KG entities as well as align or augment KG embeddings with text descriptions. Different from prior work that plugs formal entity descriptions declared in knowledge bases, this framework leverages a retriever model to selectively identify richer or highly relevant text descriptions to use in augmenting entities. Furthermore, the framework treats the number of descriptions to use in augmentation process as a parameter, which allows the flexibility of enumerating across several numbers before identifying an appropriate number. Experiment results for Link Prediction demonstrate a 5.5% and 3.5% percentage increase in the Mean Reciprocal Rank (MRR) and Hits@10 scores respectively, in comparison to text-enhanced knowledge graph augmentation methods using traditional CNNs.

A Novel Entity and Relation Joint Interaction Learning Approach for Entity Alignment

Entity alignment (EA) aims to find equivalent entities in knowledge graphs (KGs) from multiple data sources and is a crucial step in integrating KGs. Recent studies learn the similarity of entity embeddings by aggregating neighboring entities. However, these methods solely compare neighboring entities and do not incorporate the connected relation between an entity and its neighbors. In this paper, we propose a novel Entity and Relation joint Interaction Learning (ERIL) approach, which effectively captures the interaction between entities and relations, enhancing the precision of alignment across different KGs. Specifically, the ERIL model jointly learns the neighborhood features of entities and the spatial structure of relations to train a shared permutation matrix, capturing comprehensive associative relations within KGs. Moreover, a semi-supervised iterative framework is designed to leverage the positive interactions between entities and relations to identify more aligned entities. Extensive experiments are conducted on five benchmark datasets to demonstrate the effectiveness of ERIL compared with existing state-of-the-art EA methods. On DBP15K, our model ERIL outperforms currently available EA methods by 1.9% on Hits@10.

Higher-order GNN with Local Inflation for entity alignment

In the age of massive data, the construction of knowledge graph has increasingly become a forceful support for the downstream applications of artificial intelligence. However, the information of entities in knowledge graphs is usually incomplete, so it is urgent to supplement the relations of entities through entity alignment task. Frustratingly, the current entity alignment models are facing serious challenges. First, some models only focus on structural features and other auxiliary information (e.g., attributes, images and descriptions), but ignore the features of the entity itself can be scaled resulting in over-smoothing issue. Second, most models utilize higher-order networks to aggregate neighborhood information by stacking layers, but the training cost of these models are drastically higher. Third, most models are supervised or semi-supervised, but there are few pre-aligned seeds for alignment, which greatly limits the improvement of model performance. Hence, to address the above three issues, we propose a Higher-Order Graph Neural Network with Local Inflation for entity alignment, named HOLI-GNN. Specifically, we introduce a local inflation mechanism, which enlarges the each feature of entities to mitigate the impact of over-smoothing caused by neighborhood aggregation. Additionally, we propose a novel higher-order encoder to capture higher-order information. Furthermore, our model also employ currently popular iteration strategy to increase labeled entity pairs, which can markedly promote the performance of align task. Finally, we perform comprehensive experiments to validate the effectiveness of our model on benchmark datasets. The results strongly indicate that our model exhibits better performance than the state-of-the-art models.

Link Prediction Based on Feature Mapping and Bi-Directional Convolution

A considerable amount of research on link prediction has recently been driven by missing relationships between knowledge graph entities and the problem of the incompleteness of knowledge graphs. Some recent studies have shown that convolutional neural networks based on knowledge embeddings are highly expressive and have good performance in link prediction. However, we found that the convolutional neural network (CNN)-based models do not handle the link between relations and entities well. For this reason, this paper proposes a link prediction model (LPM) based on feature mapping and bi-directional convolution. For the modeling of the task, an encoding layer–mapping layer–decoding layer structure is used. Among these layers, the encoding layer adopts a graph attention network to encode multi-hop triad information and obtains richer encoding of entities and relationships. The mapping layer can realize the mapping transformation between entities and relations and project the entity encoding in the space of relation encoding to capture the subtle connection between entities and relations. The decoding layer adopts bidirectional convolution to merge and decode the triples in a sequential inverse order, which makes the decoding layer model more advantageous in prediction. In addition, the decoding layer also adopts the r-drop training method to effectively reduce the distribution error generated by training between models and enhance the robustness of the model. Our experiments demonstrated the effectiveness of mapping relations, bidirectional convolution, and r-drop, and the accuracy of the proposed model showed significant improvements for each evaluation metric on two datasets, WN18RR and FB15k-237.

SimKG-BERT: A Security Enhancement Approach for Healthcare Models Consisting of Fusing SimBERT and a Knowledge Graph

Model robustness is an important index in medical cybersecurity, and hard-negative samples in electronic medical records can provide more gradient information, which can effectively improve the robustness of a model. However, hard negatives pose difficulties in terms of their definition and acquisition. To solve these problems, a data augmentation approach consisting of fusing SimBERT and a knowledge graph for application to a hard-negative sample is proposed in this paper. Firstly, we selected 40 misdiagnosed cases of diabetic complications as the original data for data augmentation. Secondly, we divided the contents of the electronic medical records into two parts. One part consisted of the core disease phrases in the misdiagnosed case records, which a medical specialist selected. These denoted the critical diseases that the model diagnosed as negative samples. Based on these core symptom words, new symptom phrases were directly generated using the SimBERT model. On the other hand, the noncore phrases of misdiagnosed medical records were highly similar to the positive samples. We determined the cosine similarity between the embedding vector of the knowledge graph entities and a vector made up of the noncore phrases. Then, we used Top-K sampling to generate text. Finally, combining the generated text from the two parts and the disturbed numerical indexes resulted in 160 enhancement samples. Our experiment shows that the distances between the samples generated using the SimKG-BERT model’s samples were closer to those of the positive samples and the anchor points in the space vector were closer than those created using the other models. This finding is more in line with how hard negatives are defined. In addition, compared with the model without data augmentation, the F1 values in the three data sets of diabetic complications increased by 6.4%, 2.24%, and 5.54%, respectively. The SimKG-BERT model achieves data augmentation in the absence of misdiagnosed medical records, providing more gradient information to the model, which not only improves the robustness of the model but also meets the realistic needs of assisted-diagnosis safety.

Retracted: Knowledge Graph Entity Similarity Calculation under Active Learning

Retracted: Knowledge Graph Entity Similarity Calculation under Active Learning

Improving conversational recommender systems via multi-preference modelling and knowledge-enhanced

Conversational recommender systems (CRS) aim to capture user preferences and provide high-quality recommendations through conversations. Existing CRSs use items in conversations to represent user preferences, and these items usually need to match corresponding entities in external knowledge graphs (KGs). However, current general approaches extract subgraphs from large KGs (e.g., DBpedia) and thus may contain noise from nonrelevant domains and do not cover items in the CRS domain well. In addition, existing CRSs do not focus on hidden semantic preferences in user conversations, nor do they emphasize item attributes. Moreover, the system-generated responses lack diversity and descriptiveness. To address these issues, we propose a novel approach for Improving Conversation Recommendation Systems via Multi-Preference Modelling and Knowledge-Enhanced (MPKE). Specifically, we first customize a higher-quality knowledge graph LMKG for CRS to help the model fully exploit external knowledge to improve its performance. Second, we design a semantic fusion module to augment the semantic connections between external knowledge and dialogue text. Then, we propose a two-stage recommendation mechanism to explore more item attributes and fuse entity and text semantics to enhance user representation. For response generation, we use a word-level KG (i.e., ConceptNet) to enhance the representation of conversation keywords and construct explanation templates with slots to enhance the diversity of system responses. extensive experiments show that our approach outperforms past state-of-the-art methods. the source code will be published at https://github.com/zcy-cqut/MPKE.

Multi-view Few-shot Reasoning for Emerging Entities in Knowledge Graphs

Multi-view Few-shot Reasoning for Emerging Entities in Knowledge Graphs

KNOWNET: Guided Health Information Seeking from LLMs via Knowledge Graph Integration.

The increasing reliance on Large Language Models (LLMs) for health information seeking can pose severe risks due to the potential for misinformation and the complexity of these topics. This paper introduces KNOWNET a visualization system that integrates LLMs with Knowledge Graphs (KG) to provide enhanced accuracy and structured exploration. Specifically, for enhanced accuracy, KNOWNET extracts triples (e.g., entities and their relations) from LLM outputs and maps them into the validated information and supported evidence in external KGs. For structured exploration, KNOWNET provides next-step recommendations based on the neighborhood of the currently explored entities in KGs, aiming to guide a comprehensive understanding without overlooking critical aspects. To enable reasoning with both the structured data in KGs and the unstructured outputs from LLMs, KNOWNET conceptualizes the understanding of a subject as the gradual construction of graph visualization. A progressive graph visualization is introduced to monitor past inquiries, and bridge the current query with the exploration history and next-step recommendations. We demonstrate the effectiveness of our system via use cases and expert interviews.