Joint Entity Research Articles

Candidate Set Expansion for Entity and Relation Linking Based on Mutual Entity–Relation Interaction

Entity and relation linking are the core tasks in knowledge base question answering (KBQA). They connect natural language questions with triples in the knowledge base. In most studies, researchers perform these two tasks independently, which ignores the interplay between the entity and relation linking. To address the above problems, some researchers have proposed a framework for joint entity and relation linking based on feature joint and multi-attention. In this paper, based on their method, we offer a candidate set generation expansion model to improve the coverage of correct candidate words and to ensure that the correct disambiguation objects exist in the candidate list as much as possible. Our framework first uses the initial relation candidate set to obtain the entity nodes in the knowledge graph related to this relation. Second, the filtering rule filters out the less-relevant entity candidates to obtain the expanded entity candidate set. Third, the relation nodes directly connected to the nodes in the expanded entity candidate set are added to the initial relation candidate set. Finally, a ranking algorithm filters out the less-relevant relation candidates to obtain the expanded relation candidate set. An empirical study shows that this model improves the recall and correctness of the entity and relation linking for KBQA. The candidate set expansion method based on entity–relation interaction proposed in this paper is highly portable and scalable. The method in this paper considers the connections between question subgraphs in knowledge graphs and provides new ideas for the candidate set expansion.

Design knowledge graph-aided conceptual product design approach based on joint entity and relation extraction

Design knowledge is critical to creating ideas in the conceptual design stage of product development for innovation. Fragmentary design data, massive multidisciplinary knowledge call for the development of a novel knowledge acquisition approach for conceptual product design. This study proposes a Design Knowledge Graph-aided (DKG-aided) conceptual product design approach for knowledge acquisition and design process improvement. The DKG framework uses a deep-learning algorithm to discover design-related knowledge from massive fragmentary data and constructs a knowledge graph for conceptual product design. The joint entity and relation extraction model is proposed to automatically extract design knowledge from massive unstructured data. The feasibility and high accuracy of the proposed design knowledge extraction model were demonstrated with experimental comparisons and the validation of the DKG in the case study of conceptual product design inspired by massive real data of porcelain.

Dynamic Modeling Cross-Modal Interactions in Two-Phase Prediction for Entity-Relation Extraction.

Joint extraction of entities and their relations benefits from the close interaction between named entities and their relation information. Therefore, how to effectively model such cross-modal interactions is critical for the final performance. Previous works have used simple methods, such as label-feature concatenation, to perform coarse-grained semantic fusion among cross-modal instances but fail to capture fine-grained correlations over token and label spaces, resulting in insufficient interactions. In this article, we propose a dynamic cross-modal attention network (CMAN) for joint entity and relation extraction. The network is carefully constructed by stacking multiple attention units in depth to dynamic model dense interactions over token-label spaces, in which two basic attention units and a novel two-phase prediction are proposed to explicitly capture fine-grained correlations across different modalities (e.g., token-to-token and label-to-token). Experiment results on the CoNLL04 dataset show that our model obtains state-of-the-art results by achieving 91.72% F1 on entity recognition and 73.46% F1 on relation classification. In the ADE and DREC datasets, our model surpasses existing approaches by more than 2.1% and 2.54% F1 on relation classification. Extensive analyses further confirm the effectiveness of our approach.

Span-based single-stage joint entity-relation extraction model.

Extracting entities and relations from the unstructured text has attracted increasing attention in recent years. The existing work has achieved considerable results, yet it is difficult to solve entity overlap and exposure bias. To address cascading errors, exposure bias, and entity overlap in existing entity relation extraction approaches, we propose a joint entity relation extraction model (SMHS) based on a span-level multi-head selection mechanism, transforming entity relation extraction into a span-level multi-head selection problem. Our model uses span-tagger and span-embedding to construct span semantic vectors, utilizes LSTM and multi-head self-attention mechanism for span feature extraction, multi-head selection mechanism for span-level relation decoding, and introduces span classification task for multi-task learning to decode out the relation triad in a single-stage. Experiments on the classic English dataset NYT and the publicly available Chinese relationship extraction dataset DuIE 2.0 show that this method achieves better results than the baseline method, which verifies the effectiveness of this method. Source code and data are published here(https://github.com/Beno-waxgourd/NLP.git).

Joint entity–relation extraction for natural disaster based on table filling

In this article, we propose a joint extraction of entity–relation triplets in natural disaster cases based on word pair relation table filling. For computational accuracy concerns or other reasons, traditional works often do entity recognition and relation extraction separately; it might put less attention over the task connections and triplet global association. We propose the Global Table Attention GRU (GL-TGRU) model as a joint approach that uses sequence information encoding and table information encoding to jointly learn the representation and enhance the global association of entity and relation in table filling. We evaluated the proposed model on the public data set SciERC and the natural disaster data set SSD-HDS, respectively. The F1 scores of experimental results of the GL-TGRU model for entity identification and relation extraction achieved 65.81% and 37.30% on SciERC and achieved 93.89% and 84.06% on SSD-HDS. The results show our model helps to capture more global association relations of entity and relation, which can better identify the entity–relation triplet information.

A Two-Phase Paradigm for Joint Entity-Relation Extraction

An exhaustive study has been conducted to investigate span-based models for the joint entity and relation extraction task. However, these models sample a large number of negative entities and negative relations during the model training, which are essential but result in grossly imbalanced data distributions and in turn cause suboptimal model performance. In order to address the above issues, we propose a two-phase paradigm for the span-based joint entity and relation extraction, which involves classifying the entities and relations in the first phase, and predicting the types of these entities and relations in the second phase. The two-phase paradigm enables our model to significantly reduce the data distribution gap, including the gap between negative entities and other entities, as well as the gap between negative relations and other relations. In addition, we make the first attempt at combining entity type and entity distance as global features, which has proven effective, especially for the relation extraction. Experimental results on several datasets demonstrate that the span-based joint extraction model augmented with the two-phase paradigm and the global features consistently outperforms previous state-of-the-art span-based models for the joint extraction task, establishing a new standard benchmark. Qualitative and quantitative analyses further validate the effectiveness the proposed paradigm and the global features.

JoinER-BART: Joint Entity and Relation Extraction With Constrained Decoding, Representation Reuse and Fusion

JoinER-BART: Joint Entity and Relation Extraction With Constrained Decoding, Representation Reuse and Fusion

Multirelational Tensor Graph Attention Networks for Knowledge Fusion in Smart Enterprise Systems

Augmented Intelligence of Things empowered by knowledge graph drives cognitive intelligence for smart enterprise management systems (EMS). Knowledge fusion technology can effectively integrate knowledge from different sources, thereby improving the accuracy and richness of the knowledge graph, which is of great significance to the sustainable development of smart EMS. Traditional machine learning methods on graphs face challenges in the fusion of complex and multirelational enterprise knowledge graphs due to inherent defects in relation semantic and local structure information capturing. In order to break through these limitations and improve EMS knowledge graphs, we propose tensor-based graph attention networks for multirelational graph representation learning (MR-GAT), and apply it to the critical tasks in knowledge fusion: Entity and relation alignment. Specifically, we innovatively adopt tensor operations to adequately model the interactions between entities and relations in EMS knowledge graph to learn more accurate representations. Additionally, we propose a relation attention mechanism, which focuses on assigning weights in the process of aggregating local semantic information for relation learning in an EMS knowledge graph. Furthermore, we develop a joint entity and relation alignment framework by utilizing the proposed multirelational graph attention networks to improve the accuracy of knowledge fusion. Experimental evaluations on three datasets present that the proposed approach outperforms the baseline models by about 1.4% on average in terms of the mean reciprocal rank metric, which demonstrates the superior ability of the proposed MR-GAT in representation learning for knowledge fusion in smart EMS.

Road-Related Information Mining from Social Media Data: A Joint Relation Extraction and Entity Recognition Approach

Social media data have been gradually regarded as a prospective social sensor in the transportation domain for capturing road conditions. Most existing social media data-based sensors (SMDbSs) of road conditions, however, rely heavily on lexicon-based methods for information extraction and provide coarse-grained location information. Hence, this work newly devises an SMDbS based on joint relation extraction and entity recognition for sensing road conditions from social media data, which eliminates the reliance on lexicon-based methods and offers finer-grained location information in comparison with existing SMDbSs. This SMDbS development consists of four major steps, including data collection and annotation, data cleansing, two-stage information extraction, and model verification. A tweet dataset in Lexington city is exploited to demonstrate this SMDbS, which shows satisfactory information extraction performance. This study would help facilitate social media data to be an extra information source in the transportation domain.

Entity Factor: A Balanced Method for Table Filling in Joint Entity and Relation Extraction

The knowledge graph is an effective tool for improving natural language processing, but manually annotating enormous amounts of knowledge is expensive. Academics have conducted research on entity and relation extraction techniques, among which, the end-to-end table-filling approach is a popular direction for achieving joint entity and relation extraction. However, once the table has been populated in a uniform label space, a large number of null labels are generated within the array, causing label-imbalance problems, which could result in a tendency of the model’s encoder to predict null labels; that is, model generalization performance decreases. In this paper, we propose a method to mitigate non-essential null labels in matrices. This method utilizes a score matrix to calculate the count of non-entities and the percentage of non-essential null labels in the matrix, which is then projected by the power of natural constant to generate an entity-factor matrix. This is then incorporated into the scoring matrix. In the back-propagation process, the gradient of non-essential null-labeled cells in the entity factor layer is affected and shrinks, the amplitude of which is related to the size of the entity factor, thereby reducing the feature learning of the model for a large number of non-essential null labels. Experiments with two publicly available benchmark datasets show that the incorporation of entity factors significantly improved model performance, especially in the relation extraction task, by 1.5% in both cases.

A Span-based Multi-Modal Attention Network for joint entity-relation extraction

Joint extraction of entities and their relations not only depends on entity semantics but also highly correlates with contextual information and entity types. Therefore, an effective joint modelling method designed for handling information from different modalities can lead to a superior performance of the joint entity and relation extraction. Previous span-based models tended to focus on the internal semantics of a span but failed to effectively capture the interactions between the span and other modal information (such as tokens or labels). In this study, a Span-based Multi-Modal Attention Network (SMAN) is proposed for joint entity and relation extraction. The network introduces a cloze mechanism to simultaneously extract the context and span position information, and jointly models the span and label in the relation extraction stage. To determine the fine-grained associations between different modalities, a Modal-Enhanced Attention (MEA) module with two modes is designed and adopted in the modelling process. Experimental results reveal that the proposed model consistently outperforms the state-of-the-art for both entity recognition and relation extraction on the SciERC and ADE datasets, and beats other competing approaches by more than 1.42% F1 score for relation extraction on the CoNLL04 dataset. Extensive additional experiments further verify the effectiveness of the proposed model.

Specific Relation Attention-Guided Graph Neural Networks for Joint Entity and Relation Extraction in Chinese EMR

Electronic medical records (EMRs) contain a variety of valuable medical entities and their relations. The extraction of medical entities and their relations has important application value in the structuring of EMR and the development of various types of intelligent assistant medical systems, and hence is a hot issue in intelligent medicine research. In recent years, most research aims to firstly identify entities and then to recognize the relations between the entities, and often suffers from many redundant operations. Furthermore, the challenge remains of identifying overlapping relation triplets along with the entire medical entity boundary and detecting multi-type relations. In this work, we propose a Specific Relation Attention-guided Graph Neural Networks (SRAGNNs) model to jointly extract entities and their relations in Chinese EMR, which uses sentence information and attention-guided graph neural networks to perceive the features of every relation in a sentence and then to extract those relations. In addition, a specific sentence representation is constructed for every relation, and sequence labeling is performed to extract its corresponding head and tail entities. Experiments on a medical evaluation dataset and a manually labeled Chinese EMR dataset show that our model improves the performance of Chinese medical entities and relation extraction.

Semantic-consistent learning for one-shot joint entity and relation extraction

Semantic-consistent learning for one-shot joint entity and relation extraction

A Relation-Oriented Model With Global Context Information for Joint Extraction of Overlapping Relations and Entities.

The entity relation extraction in the form of triples from unstructured text is a key step for self-learning knowledge graph construction. Two main methods have been proposed to extract relation triples, namely, the pipeline method and the joint learning approach. However, these models do not deal with the overlapping relation problem well. To overcome this challenge, we present a relation-oriented model with global context information for joint entity relation extraction, namely, ROMGCJE, which is an encoder–decoder model. The encoder layer aims to build long-term dependencies among words and capture rich global context representation. Besides, the relation-aware attention mechanism is applied to make use of the relation information to guide the entity detection. The decoder part consists of a multi-relation classifier for the relation classification task, and an improved long short-term memory for the entity recognition task. Finally, the minimum risk training mechanism is introduced to jointly train the model to generate final relation triples. Comprehensive experiments conducted on two public datasets, NYT and WebNLG, show that our model can effectively extract overlapping relation triples and outperforms the current state-of-the-art methods.

OneRel: Joint Entity and Relation Extraction with One Module in One Step

Joint entity and relation extraction is an essential task in natural language processing and knowledge graph construction. Existing approaches usually decompose the joint extraction task into several basic modules or processing steps to make it easy to conduct. However, such a paradigm ignores the fact that the three elements of a triple are interdependent and indivisible. Therefore, previous joint methods suffer from the problems of cascading errors and redundant information. To address these issues, in this paper, we propose a novel joint entity and relation extraction model, named OneRel, which casts joint extraction as a fine-grained triple classification problem. Specifically, our model consists of a scoring-based classifier and a relation-specific horns tagging strategy. The former evaluates whether a token pair and a relation belong to a factual triple. The latter ensures a simple but effective decoding process. Extensive experimental results on two widely used datasets demonstrate that the proposed method performs better than the state-of-the-art baselines, and delivers consistent performance gain on complex scenarios of various overlapping patterns and multiple triples.

Extraction of Joint Entity and Relationships with Soft Pruning and GlobalPointer

In recent years, scholars have paid increasing attention to the joint entity and relation extraction. However, the most difficult aspect of joint extraction is extracting overlapping triples. To address this problem, we propose a joint extraction model based on Soft Pruning and GlobalPointer, short for SGNet. In the first place, the BERT pretraining model is used to obtain the text word vector representation with contextual information, and then the local and non-local information of the word vector is obtained through graph operations. Specifically, to address the lack of information caused by the rule-based pruning strategies, we utilize the Gaussian Graph Generator and the attention-guiding layer to construct a fully connected graph. This process is called soft pruning for short. Then, to achieve node message passing and information integration, we employ GCNs and a thick connection layer. Next, we use the GlobalPointer decoder to convert triple extraction into quintuple extraction to tackle the problem of problematic overlapping triples extraction. The GlobalPointer decoder, unlike the typical feedforward neural network (FNN), can perform joint decoding. In the end, to evaluate the model performance, the experiment was carried out on two public datasets: the NYT and WebNLG. The experiments show that SGNet performs substantially better on overlapping extraction and achieves good results on two publicly available datasets.

Joint Entity and Relation Extraction Network with Enhanced Explicit and Implicit Semantic Information

The main purpose of the joint entity and relation extraction is to extract entities from unstructured texts and extract the relation between labeled entities at the same time. At present, most existing joint entity and relation extraction networks ignore the utilization of explicit semantic information and explore implicit semantic information insufficiently. In this paper, we propose Joint Entity and Relation Extraction Network with Enhanced Explicit and Implicit Semantic Information (EINET). First, on the premise of using the pre-trained model, we introduce explicit semantics from Semantic Role Labeling (SRL), which contains rich semantic features about the entity types and relation of entities. Then, to enhance the implicit semantic information and extract richer features of the entity and local context, we adopt different Bi-directional Long Short-Term Memory (Bi-LSTM) networks to encode entities and local contexts, respectively. In addition, we propose to integrate global semantic information and local context length representation in relation extraction to further improve the model performance. Our model achieves competitive results on three publicly available datasets. Compared with the baseline model on Conll04, EINET obtains improvements by 2.37% in F1 for named entity recognition and 3.43% in F1 for relation extraction.

Boundary assembling method for joint entity and relation extraction

In recent years, a paradigm shift has occurred in the field of joint entity and relation extraction from token tagging to span classification, because the latter can handle nested named entities in a sentence and better utilize the global features of a possible named entity. Because relation extraction should verify every entity pair in a sentence, the performance of joint entity and relation extraction significantly depends on the quality of the entity span proposals. Most models enumerate numerous inaccurate entity spans, causing a severe data imbalance problem and high computational complexity. To address these problems, we propose a boundary assembling model for joint entity and relation extraction, in which entity boundaries are assembled to enumerate entity spans. Entity boundaries have small granularity and are less ambiguous, hence, the proposed model can benefit from accurate entity spans. Furthermore, boundary detection, span classification, and relation extraction are integrated into an end-to-end framework; thus, our model can share model parameters for multi-objective learning, which enhances the discriminability of a neural network. Experiments show that our boundary assembling model outperforms existing state-of-the-art models on four evaluation datasets: SciERC, ADE, ACE05, and CoNLL04.

MultiJAF: Multi-modal joint entity alignment framework for multi-modal knowledge graph

Entity Alignment (EA) is a crucial task in knowledge fusion, which aims to link entities with the same real-world identity from different Knowledge Graphs (KGs). Existing methods have achieved satisfactory performance, however, they mainly focus on single modal KG, which is difficult to be effectively applied to multi-modal scenes. In this paper, we propose a Multi-modal Joint entity Alignment Framework (MultiJAF), which can effectively utilize the knowledge of various modalities. Concretely, we first learn the embeddings of different modalities, i.e., structure, attribute and image modalities. Next, we adopt an attention-based multi-modal fusion network to integrate these embeddings and use obtained joint embeddings to compute a joint embedding-based similarity matrix SJ. Moreover, we design a Numerical Process Module (NPM) to infer a similarity matrix SN according to the numerical information of entities. In the end, we utilize a simple late fusion method to ensemble two similarity matrices for the final alignment. In addition, to reduce the cost of labeling data, we propose a novel NPM-based unsupervised multi-modal EA method. Experimental results on two real-world datasets demonstrate the effectiveness of our proposed MultiJAF.

Neural entity linking: A survey of models based on deep learning

This survey presents a comprehensive description of recent neural entity linking (EL) systems developed since 2015 as a result of the “deep learning revolution” in natural language processing. Its goal is to systemize design features of neural entity linking systems and compare their performance to the remarkable classic methods on common benchmarks. This work distills a generic architecture of a neural EL system and discusses its components, such as candidate generation, mention-context encoding, and entity ranking, summarizing prominent methods for each of them. The vast variety of modifications of this general architecture are grouped by several common themes: joint entity mention detection and disambiguation, models for global linking, domain-independent techniques including zero-shot and distant supervision methods, and cross-lingual approaches. Since many neural models take advantage of entity and mention/context embeddings to represent their meaning, this work also overviews prominent entity embedding techniques. Finally, the survey touches on applications of entity linking, focusing on the recently emerged use-case of enhancing deep pre-trained masked language models based on the Transformer architecture.