Technological developments, digital media, and advances in open government practices have made a vast amount of information available for social scientists. Most of this information is available as text. News portals disseminate political stories at unprecedented rates, politicians and political elites advertise their own messages through social media outlets and crowdsourcing provides new affordable and quick venues for asking citizens what they think about politics. With political texts at our fingertips, vexing research questions are emerging.
Extracting, organizing, and analyzing large amounts of textual information can be quite resource-intensive with many political scientists lacking the skills necessary for dealing with such data. Fortunately, recent developments of cutting edge computational technologies such as automatic language processing, machine learning, and information extraction techniques has made research utilizing text-as-data more accessible and appealing.
On the other hand, computational scholars equipped with novel technologies and linguistic solutions often have less experience with social science theories and less contextual knowledge about political data. There is a mutual benefit in connecting disparate worlds of computational text analysis and political science in analyzing political science research problems. The aim of this conference is to facilitate this multidisciplinary cooperation.
We are interested in contributions on computational approaches in analysing political text such as government speeches, political debates, social media, media content, party manifestos and/or legislation.
This conference is looking for, but not limited to, contributions from the following topics with a focus on politically relevant data:
stance classification and ideal point estimation
political reputation analysis
techniques for multilingual text analysis
other language technologies
Jon Oberlander has been Professor of Epistemics at the University of Edinburgh since 2005.
He works on getting computers to talk (or write) like individual people, so his research involves not only studying how people express themselves - face to face or online - but also building machines that can adapt themselves to people. He collaborates with linguists, psychologists, computer scientists and social scientists, and has long standing interests in the uses of technology in cultural heritage and creative industries. He was founder-Director of the Scottish Informatics and Computer Science Alliance, and is now Director of the Institute for Language, Cognition and Computation, and Co-Director of the Centre for Design Informatics.
Read more about Jon on http://homepages.inf.ed.ac.uk/jon/.
Stuart Soroka is the Michael W. Traugott Collegiate Professor of Communication Studies and Political Science, and Faculty Associate in the Center for Political Studies at the Institute for Social Research, University of Michigan. He has moved from McGill University in summer 2014.
Most of his research focuses on political communication, the sources and/or structure of public preferences for policy, and on the relationships between public policy, public opinion, and mass media.
Welcome reception and registration
Keynote: Stuart Soroka
The Potentials and Pitfalls of Automated Sentiment Analysis in Political Texts
Moderator: Sebastian Popa
Edoukou Philippe Armel N'Techobo, Amal Zouaq and Michel Gagnon: Semantic Annotation for the Analysis of Political Debates: A Graph-based Approach
Gijs Schumacher, Martijn Schoonvelde, Denise Traber, Tanushree Dahiya and Erik de Vries: EUSpeech: a New Dataset of EU Elite Speeches
Federico Nanni, Cäcilia Zirn, Goran Glavaš, Jason Eichorst and Simone Paolo Ponzetto: TopFish: Topic- Based Analysis of Political Position in US Electoral Campaigns
Andrey Kutuzov and Alexander Bai: Tracking Political Reputation with Distributional Semantic Models
Discussants: Stuart Soroka and Philip Habel
12:00-12:15 Coffee break
Moderator: Daniela Širinić
Damir Korenčić, Marijana Grbeša Zenzerović and Jan Snajder: Issues and their Salience in the 2015 Parliamentary Election in Croatia: A Topic Model based Analysis of the Media Agenda
Michael Amsler and Bruno Wueest: Legitimacy of New Forms of Governance in Public Discourse - An Automated Media Content Analysis Approach Driven by Techniques of Computational Linguistics
Blake Miller: Automated Detection of Chinese Government Astroturfers Using Network and Social Metadata
Discussants: Shaun Bevan and Dan Hiaeshutter-Rice
14:00-15:00 Lunch at Sezam
Moderator: Zoltan Fazekas
Caecilia Zirn, Goran Glavaš, Federico Nanni, Jason Eichorst and Heiner Stuckenschmidt : Classifying Topics and Detecting Topic Shifts in Political Manifestos
Zachary Greene: How Intra-Party Disagreement Determines Issue Salience in Election Manifestos
Felix Biessmann, Pola Lehmann, Daniel Kirsch and Sebastian Schelter: Predicting Political Party Affiliation from Text
Discussants: Andrej Kutuzov and Sebastian Popa
17:30 Dubrovnik Wall Tour (optional: provided by the organizers)
18:30 Dubrovnik Historic Center Tour (optional: provided by the organizers)
Keynote: Jon Oberlander
Role Theory Meets Relation Extraction: Initial Experiments on Perceptions of the EU
Moderator: Shaun Bevan
Sebastian Popa, Zoltan Fazekas, Hermann Schmitt, Pablo Barberá and Yannis Theocharis: Talk is Cheap: Selective Politicization of EU dimension in the 2014 EP elections
Dan Hiaeshutter-Rice: The Medium is the Message: Automated Content-Analytic Techniques Across Mass Media Platforms
Anjie Fang, Philip Habel, Iadh Ounis, Craig Macdonald and Xiao Yang: On Classifying Twitter Users’ Policy-Relevant Community Affiliations Using DBpedia
Discussants: Jon Oberlander and Matthew Loftis
11:30-11:45 Coffee break
Moderator: Goran Glavaš
Juraj Medzihorsky: The Mixture Index of Fit in Text Analysis
Matt W. Loftis and Peter B. Mortensen: Collaborating with the Machines: Building big political data sets on a budget
Kohei Watanabe: Computer-aided dictionary making: An efficient dictionary construction technique for content analysis
Discussants: Goran Glavaš and Zoltan Fazekas
13:30-14:30 Lunch at Sezam
Moderator: Pola Lehmann
Stefan Dahlberg , Sofia Axelsson, Markus Sahlgren and Amaru Cuba Gyllensten: The Meaning of Democracy: Using Distributional Semantics to Account for Meaning Differences
Benjamin Bagozzi, Daniel Berliner and Zack Almquist: Predicting Government (Non)Responsiveness to Freedom of Information Requests with Supervised Latent Dirichlet Allocation
Mennatallah El-Assady, Valentin Gold, Annette Hautli-Janisz, Wolfgang Jentner, Miriam Butt, Katharina Holzinger and Daniel Kleim: VisArgue - A Visual Text Analytics Framework for the Study of Deliberative Communication
Discussants: Juraj Medzihorsky and Damir Korenčić
16:15-16:30 Coffee break
Moderator: Zachary Greene
Eugenia Giraudy, Matthew Burgess, Julian Katz-Samuels and Joe Walsh: Beyond Money in Politics: Automatic Detection of Legislative Text Re-Use
Eitan Tzelgov and Petrus Sundin Olander: Mapping the Immigration Debate in the Swedish Riksdag
Discussants: Edoukou Philippe Armel N'Techobo and Randolph M. Siverson
Conference dinner for all participants at Konoba Dubrava
Meeting point: Hotel Lero and a short bus trip to the 'inlands'
Abstract submission deadline: 1 February 2016
Notification of acceptance: 15 March 2016
Registration: 10 April 2016
Short papers due: 20 June 2016
Conference: 14 - 16 July 2016
The conference will be held at the Centre for Advanced Academic Studies (CAAS), located in the very heart of Dubrovnik. CAAS is founded and managed by the University of Zagreb. CAAS conducts multidisciplinary scientific programs including postgraduate courses, international interdisciplinary workshops, scientific conferences and seminars, summer schools, TEMPUS projects, UNESCO programs and various research projects with international partners. Also, CAAS is hosting several other institutions or permanent projects and Inter - University Centre Dubrovnik (IUC) is one of them.
Accommodation will be provided for one author of each accepted paper. We encourage participation of junior scholars by providing a small number of travel grants. We will also provide assistance in finding accommodation at discount rates for co-authors.
The conference includes a complimentary reception (Thursday) and dinner (Saturday) for all participants as well as lunch and refreshments throught the conference.
All participants are kindly asked to take necessary steps for their travel arrangements soon after their participation is confirmed by the Organizing Committee.
The Dubrovnik Airport has very good connections in July with all major European airports. Alternative is to fly to Zagreb, Split or Zadar and take a bus to Dubrovnik.
Daniela Širinić - University of Zagreb
Daniela is Research Fellow at the Faculty of Political Science, University of Zagreb and a doctoral candidate in comparative politics at the Central European University. She is also the principal investigator of the Croatian Policy Agendas Project. This conference is the integral part of the project and is sponsored by the European Social Fund. Daniela’s research interests include agenda-setting, political representation, political participation, data journalism, and open data practices.
Jan Šnajder - University of Zagreb, TakeLab
Jan is an Assistant Professor at the Faculty of Electrical Engineering and Computing at the University of Zagreb and a member of Text Analysis and Knowledge Engineering Lab (TakeLab). Jan’s research interest are in natural language processing, machine learning, and language technologies. His current focus is on lexical semantics, information extraction, and opinion mining.
Shaun Bevan - University of Edinburgh
Shaun is Lecturer in Quantitative Political Science at the University of Edinburgh, an External Fellow at the Mannheim Centre for European Social Research (MZES) at the University of Mannheim and is the Topic Coding Coordinator for the Comparative Agendas Project responsible for creating a common Master Codebook across projects. Shaun's research interests include agenda-setting, public policy, interest groups, public opinion, time series analysis and measurement.
Zoltán Fazekas - University of Oslo
Zoltan is Postdoctoral Fellow in the Department of Political Science, University of Oslo. Zoltan's research focuses on political behavior at voter and elite level, with special interest in political opinion formation, contextual differences, and varied uses of social media data for the study of politics.
International Conference on the
Advances in Computational Analysis of