Deep Structured Neural Network for Event Temporal Relation Extraction
Preprint
Abstract
We propose a novel deep structured learning framework for event temporal relation extraction. The model consists of 1) a recurrent neural network (RNN) to learn scoring functions for pair-wise relations, and 2) a structured support vector machine (SSVM) to make joint predictions. The neural network automatically learns representations that account for long-term contexts to provide robust features for the structured model, while the SSVM incorporates domain knowledge such as transitive closure of temporal relations as constraints to make better globally consistent decisions. By jointly training the two components, our model combines the benefits of both data-driven learning and knowledge exploitation. Experimental results on three high-quality event temporal relation datasets (TCR, MATRES, and TB-Dense) demonstrate that incorporated with pre-trained contextualized embeddings, the proposed model achieves significantly better performances than the state-of-the-art methods on all three datasets. We also provide thorough ablation studies to investigate our model.Keywords
All Related Versions
This publication has 12 references indexed in Scilit:
- BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding2019
- A Multi-Axis Annotation Scheme for Event Temporal Relations2018
- LIMSI-COT at SemEval-2017 Task 12: Neural Architecture for Temporal Information Extraction from Clinical Narratives2017
- A Structured Learning Approach to Temporal Relation Extraction2017
- Dense Event Ordering with a Multi-Pass ArchitectureTransactions of the Association for Computational Linguistics, 2014
- Glove: Global Vectors for Word Representation2014
- An Annotation Framework for Dense Event Ordering2014
- Training structural SVMs when exact inference is intractable2008
- Jointly combining implicit constraints improves temporal ordering2008
- Learning by Reading by Learning to Read2007