Argumentation Mining on Essays at Multi Scales

Proceedings of the 28th International Conference on Computational Linguistics, pages 5480–5493 Barcelona, Spain (Online), December 8-13, 2020 5480 Argumentation Mining on Essays at Multi Scales Hao Wang1∗, Zhen Huang2∗, Yong Dou3, Yu Hong4 1,2,3 School of Computer Science, National University of Defense Technology, Changsha, Hunan, China 4 School of Computer Science and Technology, Soochow University, Suzhou, Jiangsu, China 1,2,3{hao.wang, huangzhen, yongdou}@nudt.edu.cn, 4{tianxianer}@gmail.com Abstract Argumentation mining on essays is a new challenging task in natural language processing, which aims to identify the types and locations of argumentation components. Recent research mainly models the task as a sequence tagging problem and deal with all the argumentation components at word level. However, this task is not scale-independent. Some types of argumentation compo- nents which serve as core opinions on essays or paragraphs, are at essay level or paragraph level. Sequence tagging method conducts reasoning by local context words, and fails to effectively mine these components. To this end, we propose a multi-scale argumentation mining model, where we respectively mine different types of argumentation components at corresponding lev- els. Besides, an effective coarse-to-fine argumentation fusion mechanism is proposed to further improve the performance. We conduct a serial of experiments on the Persuasive Essay dataset (PE 2.0). Experimental results indicate that our model outperforms existing models on mining all types of argumentation components. 1 Introduction Argumentation mining (AM) is a challenging task in natural language processing (Lippi and Torroni, 2016). Recent research mainly involves independent sentences (Habernal and Gurevych, 2016; Bar- Haim et al., 2017; Daxenberger et al., 2017; Niven and Kao, 2019; Reimers et al., 2019) and also essays (Levy et al., 2014; Rinott et al., 2015; Rinott et al., 2015; Lippi and Torroni, 2016; Habernal and Gurevych, 2017; Eger et al., 2017; Chernodub et al., 2019; Petasis, 2019). In this paper, we focus on argumentation mining on essays. Argumentation mining on essays aims to identify the types and locations of argumentation components from essay text (Lippi and Torroni, 2016). Typically, there are three argumentation types, namely major claims (MC), claim (C) and premises (P). Previous research (Levy et al., 2014; Rinott et al., 2015) takes sentences as the smallest argumentative unit, and handles this task in a rough way. They firstly split the essay into several sentences, and adopt a sentence classification model to select and reserve sentences which may be promising to contain argu- mentation components. Then they further identify the exact boundaries of argumentation components in those sentences. These pipeline approaches fail to conduct effective argumentation mining, since they ignore the argumentation structure of the essay and only handle the task at sentence level. Recent research (Eger et al., 2017; Chernodub et al., 2019) focuses on end-to-end neural models. They boil the task down to a sequence tagging problem, and handle it at word level instead of sentence level. Typically, neural network is employed as encoder for text representation, and Conditional Random Field (CRF) is employed as decoder to make final prediction. This word-level sequence tagging method can simultaneously identify the …

Content truncated to 500 words