A Lexicon-Based Graph Neural Network for Chinese NER

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 1040–1050, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 1040 A Lexicon-Based Graph Neural Network for Chinese NER Tao Gui∗, Yicheng Zou∗, Qi Zhang, Minlong Peng, Jinlan Fu, Zhongyu Wei, Xuanjing Huang Shanghai Key Laboratory of Intelligent Information Processing, Fudan University School of Computer Science, Fudan University 825 Zhangheng Road, Shanghai, China {tgui16,yczou18,qz,mlpeng16,fujl16,zywei,xjhuang}@fudan.edu.cn Abstract Recurrent neural networks (RNN) used for Chinese named entity recognition (NER) that sequentially track character and word infor- mation have achieved great success. How- ever, the characteristic of chain structure and the lack of global semantics determine that RNN-based models are vulnerable to word ambiguities. In this work, we try to alleviate this problem by introducing a lexicon-based graph neural network with global seman- tics, in which lexicon knowledge is used to connect characters to capture the local composition, while a global relay node can capture global sentence semantics and long- range dependency. Based on the multiple graph-based interactions among characters, potential words, and the whole-sentence se- mantics, word ambiguities can be effectively tackled. Experiments on four NER datasets show that the proposed model achieves sig- nificant improvements against other baseline models. 1 Introduction The task of named entity recognition (NER) involves determining entity boundaries and rec- ognizing categories of named entities, which is a fundamental task in the field of natural language processing (NLP). NER plays an important role in many downstream NLP tasks, including in- formation retrieval (Chen et al., 2015b), relation extraction (Bunescu and Mooney, 2005), question answering systems (Diefenbach et al., 2018), and other applications. Compared with English NER, Chinese named entities are more difficult to identify due to their uncertain boundaries, complex composition, and NE definitions within the nest (Duan and Zheng, 2011). One intuitive way to alleviate word boundary problems is to first perform word segmentation ∗Equal contribution. 印 Yin 度 Du 河 River 流 Flow 经 Through 印 Yin 度 Du 印度 India 印度河 The Indus River 河流 River 流经 Flow through 印度 India E-GPE M-LOC ? Partial Sequence B-GPE B-LOC ? Figure 1: Example of word character lattice with partial input. Because of the characteristic of chain structure, RNN-based methods must predict the label “度” using only previous partial sequences “印度 (India)”, which may suffer from word ambiguities without global sentence semantics. and then apply word sequence labeling (Yang et al., 2016; He and Sun, 2017). However, the rare gold-standard segmentation in NER datasets and incorrectly segmented entity boundaries both negatively impact the identification of named entities (Peng and Dredze, 2015; He and Sun, 2016). Hence, character-level Chinese NER using lexicon features to better leverage word informa- tion has attracted research attention (Passos et al., 2014; Zhang and Yang, 2018). In particular, Zhang and Yang (2018) introduced a variant of a long short-term memory network (lattice- structured LSTM) that encodes all potential words matching a sentence to exploit explicit word information, achieving …

Content truncated to 500 words