Document-level Event Extraction via Heterogeneous Graph-based Interaction Model with a Tracker

Document-level event extraction aims to recognize event information from a whole piece of article. Existing methods are not effective due to two challenges of this task: a) the target event arguments are scattered across sentences; b) the correlation among events in a document is non-trivial to model. In this paper, we propose Heterogeneous Graph-based Interaction Model with a Tracker (GIT) to solve the aforementioned two challenges. For the first challenge, GIT constructs a heterogeneous graph interaction network to capture global interactions among different sentences and entity mentions. For the second, GIT introduces a Tracker module to track the extracted events and hence capture the interdependency among the events. Experiments on a largescale dataset (Zheng et al., 2019) show GIT outperforms the previous methods by 2.8 F1. Further analysis reveals GIT is effective in extracting multiple correlated events and event arguments that scatter across the document. Our code is available at https://github. com/RunxinXu/GIT.

[1]  Haoran Yan,et al.  Event Detection with Multi-Order Graph Convolution and Aggregated Attention , 2019, EMNLP.

[2]  George A. Miller,et al.  WordNet: A Lexical Database for English , 1995, HLT.

[3]  Wei Xu,et al.  Doc2EDAG: An End-to-End Document-level Framework for Chinese Financial Event Extraction , 2019, EMNLP.

[4]  Xinya Du,et al.  Document-level Event-based Extraction Using Generative Template-filling Transformers , 2020, ArXiv.

[5]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[6]  Claire Cardie,et al.  Event Extraction by Answering (Almost) Natural Questions , 2020, EMNLP.

[7]  Yue Zhao,et al.  Document Embedding Enhanced Event Detection with Hierarchical and Supervised Attention , 2018, ACL.

[8]  Xinya Du,et al.  Document-Level Event Role Filler Extraction using Multi-Granularity Contextualized Encoding , 2020, ACL.

[9]  Yang Li,et al.  Event Detection without Triggers , 2019, NAACL.

[10]  Jian Liu,et al.  Exploiting the Ground-Truth: An Adversarial Imitation Based Knowledge Distillation Approach for Event Detection , 2019, AAAI.

[11]  Dongsheng Li,et al.  Exploring Pre-trained Language Models for Event Extraction and Generation , 2019, ACL.

[12]  Beth Sundheim,et al.  Overview of the Fourth Message Understanding Evaluation and Conference , 1992, MUC.

[13]  Tom M. Mitchell,et al.  Joint Extraction of Events and Entities within a Document Context , 2016, NAACL.

[14]  Dongyan Zhao,et al.  Scale Up Event Extraction Learning via Automatic Training Data Generation , 2017, AAAI.

[15]  Luca Antiga,et al.  Automatic differentiation in PyTorch , 2017 .

[16]  Xiao Liu,et al.  Jointly Multiple Events Extraction via Attention-based Graph Information Aggregation , 2018, EMNLP.

[17]  Jun Zhao,et al.  Collective Event Detection via a Hierarchical and Bias Tagging Networks with Gated Multi-level Attention Mechanisms , 2018, EMNLP.

[18]  Juan-Zi Li,et al.  Improving Event Detection via Open-domain Trigger Knowledge , 2020, ACL.

[19]  Jun Zhao,et al.  Event Extraction via Dynamic Multi-Pooling Convolutional Neural Networks , 2015, ACL.

[20]  Ralph Grishman,et al.  Joint Event Extraction via Recurrent Neural Networks , 2016, NAACL.

[21]  Jason Weston,et al.  Memory Networks , 2014, ICLR.

[22]  Yang Xiao,et al.  DCFEE: A Document-level Chinese Financial Event Extraction System based on Automatically Labeled Training Data , 2018, ACL.

[23]  Samy Bengio,et al.  Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks , 2015, NIPS.

[24]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.