Event Detection with Trigger-Aware Lattice Neural Network

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 347–356, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 347 Event Detection with Trigger-Aware Lattice Neural Network Ning Ding1,2∗, Ziran Li1,2∗, Zhiyuan Liu2,3,4, Hai-Tao Zheng1,2†, Zibo Lin1,2 1Tsinghua Shenzhen International Graduate School, Tsinghua University 2Department of Computer Science and Technology, Tsinghua University, Beijing, China 3Institute for Artificial Intelligence, Tsinghua University, Beijing, China 4State Key Lab on Intelligent Technology and Systems, Tsinghua University, Beijing, China {dingn18}@mails.tsinghua.edu.cn Abstract Event detection (ED) aims to locate trigger words in raw text and then classify them into correct event types. In this task, neural net- work based models became mainstream in re- cent years. However, two problems arise when it comes to languages without natural delim- iters, such as Chinese. First, word-based mod- els severely suffer from the problem of word- trigger mismatch, limiting the performance of the methods. In addition, even if trigger words could be accurately located, the ambi- guity of polysemy of triggers could still af- fect the trigger classification stage. To ad- dress the two issues simultaneously, we pro- pose the Trigger-aware Lattice Neural Net- work (TLNN). (1) The framework dynami- cally incorporates word and character informa- tion so that the trigger-word mismatch issue can be avoided. (2) Moreover, for polysemous characters and words, we model all senses of them with the help of an external linguistic knowledge base, so as to alleviate the prob- lem of ambiguous triggers. Experiments on two benchmark datasets show that our model could effectively tackle the two issues and outperforms previous state-of-the-art methods significantly, giving the best results. The source code of this paper can be obtained from https://github.com/thunlp/TLNN. 1 Introduction Event Detection (ED) is a pivotal part of Event Extraction, which aims to detect the position of event triggers in raw text and classify them into corresponding event types. Conventionally, the stage of locating trigger words is known as Trig- ger Identification (TI), and the stage of clas- sifying trigger words into particular event types ∗ indicates equal contribution † Corresponding author: Hai-Tao Zheng. ( E-mail: zheng.haitao@sz.tsinghua.edu.cn ) (a) An example of trigger-word mismatch. (b) An example of polysemous trigger. Figure 1: Examples about trigger-word mismatch and polysemous trigger in Event Detection. is called Trigger Classification (TC). Although neural network methods have achieved significant progress in event detection (Nguyen and Grish- man, 2015; Chen et al., 2015; Zeng et al., 2016), both steps are still exposed to the following two issues. In the TI stage, the problem of trigger-word mismatch could severely impact the performance of event detection systems. Because in languages without natural delimiters such as Chinese, main- stream approaches are mostly word-based models, in which the segmentation should be firstly per- formed as a necessary preprocessing step. Unfor- tunately, these word-wise methods neglect an im- portant problem that a trigger could be a specific part of one word or contain multiple words. As shown in Figure …

Content truncated to 500 words