Predicting Party Affiliations from European Parliament Debates

Abstract

This paper documents an ongoing effort to assess whether party group affiliation of participants in European Parliament debates can be automatically predicted on the basis of the content of their speeches, using a support vector machine multi-class model. The work represents a joint effort between researchers within Political Science and Language Technology.

The Talk of Norway: a richly annotated corpus of the Norwegian parliament, 1998–2016

“…These results compare favorably to previously published results for multi-party systems (Høyland et al 2014). We are not aware of any inter-annotator agreement studies for party classification, making it hard to compare classification scores to human performance.…”

“…Yu et al (2008) find that training an ideology classifier is possible and fairly generalizable based on their classification results on congressional speeches in the US. Høyland et al (2014), by using a similar approach, classify party affiliation in the European Parliament. While the results are generally less accurate, mostly because of the multi-party setting (in contrast to the two-party system of the US, where a majority baseline would yield results that are comparable to the best reported EU classifier configuration), they also demonstrate that some parties are harder to classify than others.…”

“…Talk of Europe (ToE;van Aggelen et al 2017) collects debates from the European Parliament. This initiative builds on the data studied in Høyland et al (2014), and makes it available in the form of an RDF graph that connects it with additional metadata on the speakers and other facets of European politics. (Meurer 2012).…”

The politics of justification? Applying the ‘Discourse Quality Index’ to the study of the European Parliament

“…Note too that the finding in the literature that MEPs from new member states quickly adapted to the voting behaviours of their party groups after 2004 (Hix et al ., 2009) does not necessarily mean that they adapted as quickly to the EP's deliberative practices. Indeed, Hoyland and Godbout (2008) show that a divergence between deliberative and voting behaviours is a general feature of the EP. By comparing a computational linguistic analysis of plenary debates with voting behaviour in roll calls, they show that MEPs often express views that differ from the rest of their groups during debates while voting with their groups in roll calls themselves.…”

National parties and group membership in the European Parliament: ideology or pragmatism?

“…Furthermore, a proposal to reform the 1976 Electoral Act by MEP Duff suggests the introduction of an additional EU-wide constituency, from which 25 MEPs would be elected on transnational lists. These institutional changes seem to favour the consolidation of the extra-parliamentary parties (see Bardi et al, 2010) studies has begun to fill in this lacuna (Klingemann et al, 2007:27-56;Hoyland, Godbout, 2008 The rest of this paper is structured as follows. Section II provides background information on the political Groups in the 2009-2014 EP.…”