Detecting Perspectives in Political Debates

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 1573–1582 Copenhagen, Denmark, September 7–11, 2017. c⃝2017 Association for Computational Linguistics Detecting Perspectives in Political Debates David Vilares Universidade da Coru˜na Departamento de Computaci´on Campus de Elvi˜na s/n, 15071 A Coru˜na, Spain david.vilares@udc.es Yulan He School of Engineering and Applied Science Aston University United Kingdom y.he@cantab.net Abstract We explore how to detect people’s per- spectives that occupy a certain proposi- tion. We propose a Bayesian modelling approach where topics (or propositions) and their associated perspectives (or view- points) are modeled as latent variables. Words associated with topics or perspec- tives follow different generative routes. Based on the extracted perspectives, we can extract the top associated sentences from text to generate a succinct summary which allows a quick glimpse of the main viewpoints in a document. The model is evaluated on debates from the House of Commons of the UK Parliament, reveal- ing perspectives from the debates without the use of labelled data and obtaining bet- ter results than previous related solutions under a variety of evaluations. 1 Introduction Stance classification is binary classification to de- tect whether people is supporting or against a topic. Existing approaches largely rely on labelled data collected under specific topics for learning su- pervised classifiers for stance classification (Mo- hammad et al., 2016a). At most time, apart from detecting one’s stance, we are interested in find- ing out the arguments behind the person’s posi- tion. Perspectives, that state people’s ideas or the facts known to one, can be contrastive, i.e. to be in favour of or against something (e.g. Brexit vs Bre- main), or non-contrastive, i.e. independent discus- sions that share a common topic (e.g. unemploy- ment and migration in the context of economy). Recent years have seen increasing interests in argumentation mining which involves the auto- matic identification of argumentative structures, e.g., the claims and premises, and detection of argumentative relations between claims and premises or evidences. However, learning mod- els for argumentation mining often require text labelled with components within argumentative structures and detailed indication of argumentative relations among them. Such labelled data is ex- pensive to obtain in practice and it is also difficult to port models trained on one domain to another. We are particularly interested in detecting dif- ferent perspectives in political debates. Essen- tially, we would like to achieve somewhere in between stance classification and argumentation mining. Given a text document, we want to iden- tify a speaker’s key arguments, without the use of any labelled data. For example, in debates about ‘Education’, we want to automatically extract sen- tences summarising the key perspectives and their arguments, e.g. ‘our education system needs to promote excellence in stem subjects’, ‘teenagers need to be taught with sexual and health educa- tion’ or ‘grammar schools promote inequality’. Similarly, if ‘Brexit’ is being discussed in terms of leaving or remaining, we want to cluster argu- ments into those two viewpoints. To do this, we introduce a Latent Argument Model (LAM) …

Content truncated to 500 words