Employing a Combination of Qualitative Methods, Computational Analysis, and Natural Language Processing to Study the Structure and Language of Social Media Data Sets During Crisis Events
Abstract: During the last decade, social media have become a crucial communication channel during mass political or civic events by shaping “a civic and democratic discourse in a vacuum of opportunities” (Howard 2010). As academics debate the nature of social media as an alternative public sphere (Papacharissi 2009), it is important to study the use of social networks by the publics during crises, especially in regards to the origin of event frames (Fossato et al. 2008), news framing and distribution (Oates and Lokot 2012), and viral content. Various qualitative and quantitative approaches have been proposed to study social media activity during crisis events. While a cohort of qualitative social scientists and media scholars have relied on traditional interviews, content and discourse analysis to study social media use in activism and during political unrest (Tufekci and Wilson 2012, Haciyakupoglu and Zhang 2015), a vast number of quantitative approaches based on a mixture of natural language processing, network analysis and statistics has been also proposed by the computational social science community (Cohen and Ruths 2013, González-Bailón et al. 2011, González-Bailón et al. 2013).
The workshop seeks to cover the different aspects of studying and interrogating a collection of data from a crisis event, such as the Winter of Discontent in Russia, the Euromaidan protest or the conflict in eastern Ukraine using a combination of methods, such as qualitative textual and discourse analysis, framing analysis, network analysis, natural language processing, sentiment analysis. The aim of the workshop is to show the value of multidisciplinary research collaborations and to brainstorm productive use of mixed methods approaches to large data sets from large-scale political or social upheavals.
Pre-Workshop Prep:
No Prerequisites Required
Workshop Contacts
Dmytro Karamshuk – [email protected]
Tetyana Lokot – [email protected]
Presenter Bios:
Dmytro Karamshuk is a posdoctoral research associate at King’s College London. His research is focused on understanding patterns in digital traces of Internet users to assist design of future communication networks. He has previously worked on studying watching habits of BBC iPlayer users who collectively represent over 44% of UK’s households (collaboration with BBC R&D), mining geo-location social networks (e.g., Foursquare) for geographic retail analysis (collaboration with the University of Cambridge) and predicting behavior of users in online social curation websites (e.g., Pinterest). He can be reached at [email protected], or by Twitter at @karamshuk.
Tetyana Lokot is a doctoral candidate at the University of Maryland researching the role of digital media in protests in Ukraine and Russia. She is currently working on a dissertation about the Euromaidan protests in Ukraine and augmented dissent, with a focus on the affordances of digital media and social networks to augment protest tactics, and the meaning of ICT use in protests. Her other research interests include: digital activism and political participation online, impact of internet communities on political decisions, social media platforms and social/political change, post-Soviet online spaces, information manipulation and propaganda online, and viral media. She can be reached at [email protected], or by Twitter at @tanyalokot.
Oleksandr Pryymak is a data scientist at Facebook in London. He holds a PhD in Computer Science from the University of Southampton. He can be reached at [email protected].
Dr Jisun An is a postdoctoral researcher in the Social Computing team at Qatar Computing Research Institute (QCRI). Her research interests are in social media, news, bias, and politics. Her research lies at the intersection of machine learning, network science, social science, and human computer interaction. She can be reached at [email protected].