Christopher Cochrane (University of Toronto)
The Automated Detection of Emotion in Transcripts of Parliamentary Speech: Comparing Human and Machine Classification of the Video and Written Records of Debates in the Canadian House of Commons
You can learn more about Professor Cochrane by clicking here
Abstract: The volume of accessible, machine-readable written communication about politics has increased exponentially over the past three decades, spawning the pursuit of new tools for automated analyses of sentiment in large political corpora like Twitter, Facebook, and traditional print media. Unlike written communication, however, the expression of emotion in speech is not confined to word-choice and syntax, and instead relies primarily on the mechanisms of intonation, facial expressions, and body language, which go undetected in analyses of political text. This raises the question of whether tools developed for the analyses of writing will work to detect sentiment in transcripts of political speech. In this presentation, which is part of the Linked Parliamentary Data Project, we survey a variety of strategies for the automated analysis of emotion in text, and test their outputs against human-coded sentiment analysis of the written and video record of debates in the Canadian House of Commons.
For more information on the CSDC speaker series, please visit: http://csdc-cecd.ca/events/csdc-speaker-series/
This series is sponsored by the Inter-university Centre for the Study of Democratic Citizenship, which is funded by the Fonds de recherche du Québec – Société et culture (FRQSC).