Speculation and Negation Scope Detection via Convolutional Neural Networks

Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 815–825, Austin, Texas, November 1-5, 2016. c⃝2016 Association for Computational Linguistics Speculation and Negation Scope Detection via Convolutional Neural Networks Zhong Qian1, Peifeng Li1, Qiaoming Zhu1, Guodong Zhou1, Zhunchen Luo2 and Wei Luo2 School of Computer Science and Technology, Soochow University, Suzhou, 215006, China1 China Defense Science and Technology Information Center, Beijing, 100142, China2 qianzhongqz@163.com, {pfli, qmzhu, gdzhou}@suda.edu.cn, zhunchenluo@gmail.com, htqxjj@126.com Abstract Speculation and negation are important infor- mation to identify text factuality. In this paper, we propose a Convolutional Neural Network (CNN)-based model with probabilistic weighted average pooling to address specula- tion and negation scope detection. In particular, our CNN-based model extracts those meaning- ful features from various syntactic paths be- tween the cues and the candidate tokens in both constituency and dependency parse trees. Eval- uation on BioScope shows that our CNN-based model significantly outperforms the state-of- the-art systems on Abstracts, a sub-corpus in BioScope, and achieves comparable perfor- mances on Clinical Records, another sub- corpus in BioScope. 1 Introduction Factual information is critical to understand a sen- tence or a document in most typical NLP applica- tions. Speculation and negation extraction has been drawing more and more attentions in recent years due to its importance in distinguishing coun- terfactual or uncertain information from the facts. Generally speaking, speculation is a type of uncer- tain expression between certainty and negation, while negation is a grammatical category which reverses the truth value of a proposition. Commonly, speculation and negation extraction involves two typical subtasks: cue identification and scope detection. Here, a cue is a word or phrase that has speculative or negative meaning (e.g., suspect, guess, deny, not), while a scope is a text fragment governed by the corresponding cue in a sentence. Consider the following two sentenc- es for examples: (S1) The doctors warn that smoking [may harm our lungs]. (S2) He does [not like playing football] but likes swimming. 1 In sentence S1, the speculative cue “may” gov- erns the scope “may harm our lungs”, while the negative cue “not” governs the scope “not like playing football” in sentence S2. Previous work have achieved quite success on cue identification (e.g., with F1-score of 86.79 for speculative cue detection in Tang et al. (2010)). In comparison, speculation and negation scope detec- tion is still a challenge due to its inherent difficul- ties and those upstream errors. In this paper, we focus on scope detection. Previous work on scope detection can be classified into heuristic rules based methods (e.g., Özgür et al., 2009; Øvrelid et al., 2010), machine learning based methods (e.g., Tang et al., 2010; Zou et al., 2013), and hybrid approaches which integrate empirical models with manual rules (Velldal et al., 2012). Different from those previous studies, this paper presents a Convolutional Neural Network (CNN)- based approach for scope detection. CNN models, firstly invented to capture more abstract features for computer vision (LeCun et al., 1989), have achieved certain success on various NLP tasks in …

Content truncated to 500 words