Stance Classification of Context-Dependent Claims

Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 251–261, Valencia, Spain, April 3-7, 2017. c⃝2017 Association for Computational Linguistics Stance Classification of Context-Dependent Claims Roy Bar-Haim1, Indrajit Bhattacharya2, Francesco Dinuzzo3∗ Amrita Saha2, and Noam Slonim1 1IBM Research - Haifa, Mount Carmel, Haifa, 31905, Israel 2IBM Research - Bangalore, India 3IBM Research - Ireland, Damastown Industrial Estate, Dublin 15, Ireland {roybar,noams}@il.ibm.com, {indrajitb,amrsaha4}@in.ibm.com Abstract Recent work has addressed the problem of detecting relevant claims for a given con- troversial topic. We introduce the com- plementary task of claim stance classi- fication, along with the first benchmark dataset for this task. We decompose this problem into: (a) open-domain target identification for topic and claim (b) senti- ment classification for each target, and (c) open-domain contrast detection between the topic and the claim targets. Manual annotation of the dataset confirms the ap- plicability and validity of our model. We describe an implementation of our model, focusing on a novel algorithm for con- trast detection. Our approach achieves promising results, and is shown to out- perform several baselines, which represent the common practice of applying a single, monolithic classifier for stance classifica- tion. 1 Introduction The need for making persuasive arguments arises in many domains, including politics, law, mar- keting, and financial and business advising. On- demand generation of pro and con arguments for a given controversial topic would therefore be of great practical value. Natural use cases include debating support, where the user is presented with persuasive arguments for a topic of interest, and decision support, where the pros and cons of a given proposal are presented to the user. A notable research effort in this area is the IBM Debater® project whose goal is “to develop technologies that can assist humans to debate and ∗Present affiliation - Amazon. reason”1. As part of this research, Levy et al. (2014) have developed context-dependent claim detection. Given a controversial topic, such as (1) The sale of violent video games to minors should be banned , their system extracts, from corpora such as Wikipedia, Context-Dependent Claims (CDCs), defined as “general, concise statements that di- rectly support or contest the given Topic”. A claim forms the basis of an argument, being the assertion that the argument aims to establish, and therefore claim detection may be viewed as a first step in au- tomated argument construction. Recent research on claim detection (Levy et al., 2014; Lippi and Torroni, 2015) was facilitated by the IBM argu- mentative structure dataset (Aharoni et al., 2014), which contains manually collected claims for a va- riety of topics, as well as supporting evidence. In this work we introduce the related task of Claim Stance Classification: given a topic, and a set of claims extracted for it, determine for each claim whether it supports or contests the topic. Sorting extracted claims into Pro and Con would clearly improve the usability of both debating and decision support systems. We introduce the first benchmark for …

Content truncated to 500 words