‘Aye’ or ‘No’? Speech-level Sentiment Analysis of Hansard UK Parliamentary Debate Transcripts

‘Aye’ or ‘No’? Speech-level Sentiment Analysis of Hansard UK Parliamentary Debate Transcripts Gavin Abercrombie and Riza Batista-Navarro School of Computer Science University of Manchester Kilburn Building, Manchester M13 9PL gavin.abercrombie@postgrad.manchester.ac.uk riza.batista@manchester.ac.uk Abstract Transcripts of UK parliamentary debates provide access to the opinions of politicians towards many important topics, but due to the large quantity of textual data and the specialised language used, they are not straightforward for human readers to process. We apply opinion mining methods to these transcripts to classify the sentiment polarity of speakers as being either positive or negative towards the motions proposed in the debates. We compare classification performance on a novel corpus using both manually annotated sentiment labels and labels derived from the speakers’ votes (‘aye’ or ‘no’). We introduce a two-step classification model, and evaluate the performance of both one- and two-step models, as well as the use of a range of textual and contextual features. Results suggest that textual features are more indicative of manually annotated class labels. Conversely, in addition to boosting performance, contextual metadata features are particularly indicative of vote labels. Use of the two-step debate model results in performance gains and appears to capture some of the complexity of the debate format. Optimum performance on this data is achieved using all features to train a multi-layer neural network, indicating that such models may be most able to exploit the relationships between textual and contextual cues in parliamentary debate speeches. Keywords: Hansard transcripts, parliamentary debates, sentiment analysis 1. Introduction In the United Kingdom, transcripts of parliamentary de- bates (known as Hansard) are publicly and freely available. This provides access to a wealth of information concern- ing the opinions and attitudes of Members of Parliament (MPs) and their parties, towards arguably the most impor- tant topics facing society, as well as potential insights into the parliamentary democratic process. However, the large quantity of recorded material in Hansard, combined with the esoteric speaking style and opaque procedural language of Parliament, makes manual retrieval of information from these data a daunting task for the non-expert citizen. Despite the fact that opinion mining has been one of the most active areas of research in natural language process- ing (NLP), and a widespread need for political information has been cited as a motivation for the development of opin- ion mining technologies (Pang and Lee, 2008), automatic analysis of the positions taken by speakers in parliamen- tary debates has received relatively little attention from re- searchers. Sentiment anlysis is the task of automatically identifying the polarity (positive or negative) of the position taken by the holder of an opinion towards a target, such as an orga- nization, a policy, a movement, or a product. We apply sen- timent analysis methods to speeches made in the House of Commons of the UK Parliament to classify their sentiment polarity as being either positive (in support) or negative (in opposition) towards the target of each speech; that is, the motion proposed in the debate in question. Prior work on this …

Content truncated to 500 words