End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures

Abstract

We present a novel end-to-end neural model to extract entities and relations between them. Our recurrent neural network based model captures both word sequence and dependency tree substructure information by stacking bidirectional tree-structured LSTM-RNNs on bidirectional sequential LSTM-RNNs. This allows our model to jointly represent both entities and relations with shared parameters in a single model. We further encourage detection of entities during training and use of entity information in relation extraction via entity pretraining and scheduled sampling. Our model improves over the state-of-the-art feature-based model on end-to-end relation extraction, achieving 12.1% and 5.7% relative error reductions in F1-score on ACE2005 and ACE2004, respectively. We also show that our LSTM-RNN based model compares favorably to the state-of-the-art CNN based model (in F1-score) on nominal relation classification (SemEval-2010 Task 8). Finally, we present an extensive ablation analysis of several model components.

Accepted for publication at the Association for Computational Linguistics (ACL), 2016. 13 pages, 1 figure, 6 tables

References in corpus (4)

• Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
• When Are Tree Structures Necessary for Deep Learning of Representations?
• Classifying Relations via Long Short Term Memory Networks along Shortest Dependency Path
• Task-Oriented Learning of Word Embeddings for Semantic Relation Classification

Cited by in corpus (60)

• Joint entity recognition and relation extraction as a multi-head selection problem
• TENER: Adapting Transformer Encoder for Named Entity Recognition
• Cross-Sentence N-ary Relation Extraction with Graph LSTMs
• Joint Extraction of Entities and Relations Based on a Novel Tagging Scheme
• Classifying medical relations in clinical text via convolutional neural networks
• GraphIE: A Graph-Based Framework for Information Extraction
• Deep Neural Approaches to Relation Triplets Extraction: A Comprehensive Survey
• Joint Extraction of Entities and Relations Based on a Novel Decomposition Strategy
• Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers
• A Joint Many-Task Model: Growing a Neural Network for Multiple NLP Tasks
• Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Construction
• Learning for Biomedical Information Extraction: Methodological Review of Recent Advances
• An attentive neural architecture for joint segmentation and parsing and its application to real estate ads
• End-to-end Named Entity Recognition and Relation Extraction using Pre-trained Language Models
• Encoding Sentences with Graph Convolutional Networks for Semantic Role Labeling
• SEE: Syntax-aware Entity Embedding for Neural Relation Extraction
• An Attentive Sequence Model for Adverse Drug Event Extraction from Biomedical Text
• Denoising Distant Supervision for Relation Extraction via Instance-Level Adversarial Training
• A Hierarchical Framework for Relation Extraction with Reinforcement Learning
• Incorporating Relation Paths in Neural Relation Extraction
• Jointly Predicting Predicates and Arguments in Neural Semantic Role Labeling
• Fine-tuning BERT for Joint Entity and Relation Extraction in Chinese Medical Text
• Multi-labeled Relation Extraction with Attentive Capsule Network
• Parallel Scheduled Sampling
• HIN: Hierarchical Inference Network for Document-Level Relation Extraction
• Deep Neural Network Based Relation Extraction: An Overview
• Improving Aspect Term Extraction with Bidirectional Dependency Tree Representation
• Neural Relation Extraction via Inner-Sentence Noise Reduction and Transfer Learning
• "Is depression related to cannabis?": A knowledge-infused model for Entity and Relation Extraction with Limited Supervision
• BERT-GT: Cross-sentence n-ary relation extraction with BERT and Graph Transformer
• Neural Snowball for Few-Shot Relation Learning
• Feature Assisted bi-directional LSTM Model for Protein-Protein Interaction Identification from Biomedical Texts
• Relation Mention Extraction from Noisy Data with Hierarchical Reinforcement Learning
• Ensemble Neural Relation Extraction with Adaptive Boosting
• Distant Supervision for Relation Extraction with Linear Attenuation Simulation and Non-IID Relevance Embedding
• Contextualised Graph Attention for Improved Relation Extraction
• Deeper Task-Specificity Improves Joint Entity and Relation Extraction
• DIAG-NRE: A Neural Pattern Diagnosis Framework for Distantly Supervised Neural Relation Extraction
• Multiple Range-Restricted Bidirectional Gated Recurrent Units with Attention for Relation Classification
• BERT-Based Multi-Head Selection for Joint Entity-Relation Extraction
• Stochastic Learning of Nonstationary Kernels for Natural Language Modeling
• Type-aware Convolutional Neural Networks for Slot Filling
• Who Blames Whom in a Crisis? Detecting Blame Ties from News Articles Using Neural Networks
• A Trigger-Sense Memory Flow Framework for Joint Entity and Relation Extraction
• Syntax-Enhanced Neural Machine Translation with Syntax-Aware Word Representations
• Lexicosyntactic Inference in Neural Models
• Adversarial Learning for Supervised and Semi-supervised Relation Extraction in Biomedical Literature
• Extracting N-ary Cross-sentence Relations using Constrained Subsequence Kernel
• Uncovering Relations for Marketing Knowledge Representation
• Learning Context-Aware Representations of Subtrees
• Automatic Severity Classification of Coronary Artery Disease via Recurrent Capsule Network
• ClaiRE at SemEval-2018 Task 7 - Extended Version
• Multi-Task Learning for Argumentation Mining
• High-Fidelity Vector Space Models of Structured Data
• Indirect Supervision for Relation Extraction using Question-Answer Pairs
• A Conditional Cascade Model for Relational Triple Extraction
• Scientific Relation Extraction with Selectively Incorporated Concept Embeddings
• Approach for Semi-automatic Construction of Anti-infective Drug Ontology Based on Entity Linking
• Semantic Loss Application to Entity Relation Recognition
• Towards Time-Aware Distant Supervision for Relation Extraction