Multi-Graph Cooperative Learning Towards Distant Supervised Relation Extraction

BackgroundWith the rapid expansion of biomedical literature, biomedical information extraction has attracted increasing attention from researchers. In particular, relation extraction between 2 entities is a long-term research topic.ObjectiveThis study aimed to perform 2 multiclass relation extraction tasks of Biomedical Natural Language Processing Workshop 2019 Open Shared Tasks: relation extraction of Bacteria-Biotope (BB-rel) task and binary relation extraction of plant seed development (SeeDev-binary) task. In essence, these 2 tasks are aimed at extracting the relation between annotated entity pairs from biomedical texts, which is a challenging problem.MethodsTraditional research methods adopted feature- or kernel-based methods and achieved good performance. For these tasks, we propose a deep learning model based on a combination of several distributed features, such as domain-specific word embedding, part-of-speech embedding, entity-type embedding, distance embedding, and position embedding. The multi-head attention mechanism is used to extract the global semantic features of an entire sentence. Meanwhile, we introduced a dependency-type feature and the shortest dependency path connecting 2 candidate entities in the syntactic dependency graph to enrich the feature representation.ResultsExperiments show that our proposed model has excellent performance in biomedical relation extraction, achieving F1 scores of 65.56% and 38.04% on the test sets of the BB-rel and SeeDev-binary tasks. Especially in the SeeDev-binary task, the F1 score of our model is superior to that of other existing models and achieves state-of-the-art performance.ConclusionsWe demonstrated that the multi-head attention mechanism can learn relevant syntactic and semantic features in different representation subspaces and different positions to extract comprehensive feature representation. Moreover, syntactic dependency features can improve the performance of the model by learning dependency relation between the entities in biomedical texts.

Distant supervised relation extraction has been widely used to find novel relational facts from text. However, distant supervision inevitably accompanies with the wrong labelling problem, and these noisy data will substantially hurt the performance of relation extraction. To alleviate this issue, we propose a sentence-level attention-based model for relation extraction. In this model, we employ convolutional neural networks to embed the semantics of sentences. Afterwards, we build sentence-level attention over multiple instances, which is expected to dynamically reduce the weights of those noisy instances. Experimental results on real-world datasets show that, our model can make full use of all informative sentences and effectively reduce the influence of wrong labelled instances. Our model achieves significant and consistent improvements on relation extraction as compared with baselines.

Distantly Supervised Relation Extraction (DSRE) aims to automatically identify semantic relationships within large text corpora by aligning with external knowledge bases. Despite the success of current methods in automating data annotation, they introduce two main challenges: label noise and data long-tail distribution. Label noise results in inaccurate annotations, which can undermine the quality of relation extraction. The long-tail problem, on the other hand, leads to an imbalanced model that struggles to extract less frequent, long-tail relations. In this paper, we introduce a novel relation extraction framework based on multi-level hierarchical attention. This approach utilizes Graph Attention Networks (GATs) to model the hierarchical structure of the relations, capturing the semantic dependencies between relation types and generating relation embeddings that reflect the overall hierarchical framework. To improve the classification process, we incorporate a multi-level classification structure guided by hierarchical attention, which enhances the accuracy of both head and tail relation extraction. A local probability constraint is introduced to ensure coherence across the classification levels, fostering knowledge transfer from frequent to less frequent relations. Experimental evaluations on the New York Times (NYT) dataset demonstrate that our method outperforms existing baselines, particularly in the context of long-tail relation extraction, offering a comprehensive solution to the challenges of DSRE.

In recent years, substantial progress has been made on Graph Convolutional Networks (GCNs). However, the computing of GCN usually requires a large memory space for keeping the entire graph. In consequence, GCN is not flexible enough, especially for large scale graphs in complex real-world applications. Fortunately, for transductive graph representation learning, methods based on Matrix Factorization (MF) naturally support constructing mini-batches, and thus are more friendly to distributed computing compared with GCN. Accordingly, in this paper, we analyze the connections between GCN and MF, and simplify GCN as matrix factorization with unitization and co-training. Furthermore, under the guidance of our analysis, we propose an alternative model to GCN named Unitized and Co-training Matrix Factorization (UCMF). Extensive experiments have been conducted on several real-world datasets. On the task of semi-supervised node classification, the experimental results illustrate that UCMF achieves similar or superior performances compared with GCN. Meanwhile, distributed UCMF significantly outperforms distributed GCN methods, which shows that UCMF can greatly benefit complex real-world applications.

Boundary assembling method for joint entity and relation extraction

Relation Extraction (RE) and Entity Typing (ET) are two important tasks in natural language processing field. Existing methods for RE and ET usually handle them separately. However, relation extraction and entity typing have strong relatedness with each other, since entity types are informative for inferring relations between entities, and the relations can provide important information for predicting types of entities. Exploiting the relatedness between relation extraction and entity typing has the potential to improve the performance of both tasks. In this paper, we propose a neural network based approach to jointly train relation extraction and entity typing models using a multi-task learning framework. For relation extraction, we adopt a piece-wise Convolutional Neural Network model as sentence encoder. For entity typing, since there are multiple entities in one sentence, we design a couple-attention model based on Bidirectional Long Short-Term Memory network to obtain entity-specific representation of sentences. In our MTL frame, the two tasks share not only the low-level input embeddings but also the high-level task-specific semantic representations with each other. The experiment results on benchmark datasets demonstrate that our approach can effectively improve the performance of both relation extraction and entity typing.

A region-based hypergraph network for joint entity-relation extraction

Cross-Language Learning for Arabic Relation Extraction

Document-level relation extraction with multi-layer heterogeneous graph attention network

With the continuous improvement of the related relation extraction model, researchers focused on optimizations of the input of the sentence expression structures, such as dependency trees, and slight improvements and integrations of basic models applied to this task. This paper utilizes position-aware attention mechanisms to highlight critical information and makes good use of global information generated by the bidirectional LSTM, absorbing characteristics and advantages of those related models to propose a new Position-Aware Graph Convolutional Network (PA-GCN) model and describe it in three steps. Moreover, this paper trials the model on the RACRED dataset to evaluate its performance and suitability. The experiment results demonstrate that the PA-GCN model outperformed others listed in this paper, achieving an F_1 score of 66.6%, which verifies an improvement of the model and the effectiveness of adding an attention-aware mechanism.

We propose a distance supervised relation extraction approach for long-tailed, imbalanced data which is prevalent in real-world settings. Here, the challenge is to learn accurate "few-shot" models for classes existing at the tail of the class distribution, for which little data is available. Inspired by the rich semantic correlations between classes at the long tail and those at the head, we take advantage of the knowledge from data-rich classes at the head of the distribution to boost the performance of the data-poor classes at the tail. First, we propose to leverage implicit relational knowledge among class labels from knowledge graph embeddings and learn explicit relational knowledge using graph convolution networks. Second, we integrate that relational knowledge into relation extraction model by coarse-to-fine knowledge-aware attention mechanism. We demonstrate our results for a large-scale benchmark dataset which show that our approach significantly outperforms other baselines, especially for long-tail relations.

Distant supervision is a hotspot in relation extraction research. Instead of relying on annotated text, distant supervision hires a knowledge base as supervision. For each pair of entities that appears in some knowledge base's relation, this approach find all sentences containing those entities in a large unlabeled corpus and extract textual features to train a relation classifier. The automatic labeling provides a large amount of data, but the data have serious problem. Most features appear only few times in training data, and such insufficient data make these features very susceptible to noise, which will lead to a flawed classifier. In this paper, we propose a method to improve few occurrence features' performance in distant supervision relation extraction. We present a novel model to calculating the similarity between a feature and an entity pair, and then adjust the entity pair' features by their similarity. The experiment shows our method boosted the performance of relation extraction.

The objective of extracting relations between document components lies in identifying the Within a single document, the focus often lies on understanding the connections between different entities. This type of analysis goes beyond individual sentences, as it demands an understanding of how information from multiple sentences interacts to form these connections. Over recent years, the importance of exploring relationships involving several entities simultaneously has grown significantly. To advance the field of studying such connections across entire documents, a novel collection of data points, known as DocRED, has been introduced. Currently, the standard approach for this task involves using BiLSTM networks to process the entire document as a whole. However, this method struggles to effectively capture the intricate relationships that exist among various entities. To overcome this limitation, a new model designed for document-level relationship identification has been developed, which leverages Graph Convolutional Networks (GCN). GCNs are particularly useful here because they can gather information from surrounding entities, allowing for a more detailed modeling of their interactions. The proposed approach starts by identifying coreferential links to gather features that represent the relationships between pairs of entities. These features are then analyzed using GCN to construct a graph structure that represents the entire document, ultimately revealing the complex interactions between different entities. Testing this model on the large-scale DocRED dataset from Tsinghua University has demonstrated its strong performance in this challenging task.

Organized relational knowledge in the form of “knowledge graphs” is important for many applications. However, the ability to populate knowledge bases with facts automatically extracted from documents has improved frustratingly slowly. This paper simultaneously addresses two issues that have held back prior work. We first propose an effective new model, which combines an LSTM sequence model with a form of entity position-aware attention that is better suited to relation extraction. Then we build TACRED, a large (119,474 examples) supervised relation extraction dataset obtained via crowdsourcing and targeted towards TAC KBP relations. The combination of better supervised data and a more appropriate high-capacity model enables much better relation extraction performance. When the model trained on this new dataset replaces the previous relation extraction component of the best TAC KBP 2015 slot filling system, its F1 score increases markedly from 22.2% to 26.7%.

Multiple perspectives can be used to explore rich and complex datasets that are widely used in many applications. However, in real-world applications, the multi-view data are often noisy because of various environmental factors. The key challenge of graph-based multi-view clustering is obtaining a consistent clustering structure. Most graph-based methods learn independently in one view how similar the data points in each view are to one another. However, the consistency of information in a multi-view dataset is easily overlooked, causing an unsatisfactory unified graph matrix. To address this problem, this study proposes a novel self-weighted graph-based framework for multi-view clustering (SGMF). This framework integrates these initial similar graph matrices into a unified graph matrix by assigning different weights to the initial graphs of each view. This algorithm’s goal is to effectively learn a unified graph matrix and eliminate noise or irrelevant information from the dataset. This is the first time a generalized multi-view framework based on the L1-norm has been proposed. L1-norm can be used to improve the robustness of the algorithm. Particularly, SGMF can automatically weigh each view’s initial similar graph matrix to solve the noise problem and obtain a consistent data structure. Simultaneously, a rank constraint is imposed on the graph Laplacian matrix of the final unified matrix, which helps to automatically divide the data points into the appropriate number of clusters. The experimental results on synthetic and real datasets verify the effectiveness and superiority of the proposed method over the latest baseline.