Keynote lectures

21.08. 9:30-10:30

The talk will give an overview about some methodological shifts in relation to digital methods in social sciences and humanities. The topics will include varieties of digital data and methods, methodological limitations and innovations in relation to (big) data turn. Talk will also focus on challenges and threats related to digital data and methods, like current discussions about how to provide integrative instead of discriminatory data practices, in the context of rising automatized methods, algorithmic governance and selection processes.

Anu Masso is currently senior researcher in data studies at Institute of Social Studies at University of Tartu and visiting researcher at Chair of Social Networks at ETH Zürich. Her special passion are data and research methods in social sciences: she is interested in (1) developing measurements and estimating bias of social dynamics of large-scale socio-cultural phenomena (e.g. social network analysis methods, comparative social media analytics), and (2) in critical data studies (e.g. methodological shifts is social sciences in relation to big data turn, developing solutions for ‘data acculturation’ – culturally integrative data practices in the field of spatial mobilities). She has currently initiated a text-book project ‘Research methods for studying datafied society’, that is co-edited together with Andra Siibak and Katrin Tiidenberg and that will be published by Tallinn University Press in 2019; the book will include over 20 chapters written in Estonian by experts in the field of digital methods.

22.08. 9:30-10:30

This talk will illustrate the potential for data science on the web with examples of projects that gather and analyse large quantities of online data for a variety of research goals. The topics will include altmetrics (impact indicators for academic publications) and book genre analysis.

Mike Thelwall is Professor of Information Science and leader of the Statistical Cybermetrics Research Group at the University of Wolverhampton, which he joined in 1989. He is also Docent at the Department of Information Studies at Åbo Akademi University, and a research associate at the Oxford Internet Institute. His PhD was in Pure Mathematics from the University of Lancaster. His current research field includes identifying and analysing web phenomena using quantitative-led research methods, including altmetrics and sentiment analysis, and has pioneered an information science approach to link analysis. Mike has developed a wide range of tools for gathering and analysing web data, including hyperlink analysis, sentiment analysis and content analysis for Twitter, YouTube, MySpace, blogs and the web in general. His 400+ publications include 244 refereed journal articles and two books, including Introduction to Webometrics. He is an associate editor of the Journal of the Association for Information Science and Technology and sits on three other editorial boards. For more information, see: http://www.scit.wlv.ac.uk/~cm1993/mycv.html 

23.08. 9:30-10:30

As researchers, we are regulated by ethics oversight mechanisms, many of which waive their need to be involved if the “data” analyzed “is already public.” Alternatively, we have access to many procedural ethics guidelines. Typically, these focus on doing no harm, which is considered to be accomplishable primarily through procedures of informed consent and confidentiality. However, there is mounting critique of the insufficiency of both of these approaches. Private and public are not a binary, harm and risk can hardly be assessed in the era where technology design, platform affordances, data management, data slippage and usage practices intersect in ways that have obvious ethical implications.

This talk is about the messy, impossible and difficult situations, decisions, and what-if imaginaries when studying practices and cultures on social media. It’s based on an assumption that most of our standardized and streamlined attempts at managing risks are inadequate, while conceding to the fact that there is only so much a standardized guideline can accomplish, and acknowledging that what we have is most certainly better than nothing. I discuss how the socio-technical affordances of platforms, people’s perceptions of those, and researchers’ needs may clash as previously ephemeral aspects of human interaction seem readily available as “data;”  offers examples of decision making and dilemmas from my own studies; and blends ideas from the ethics of vulnerability, the ethics of care and moral agency to operationalize the attempts of being a decent human being as a researcher.

Katrin Tiidenberg, PhD is an Associate Professor of Social Media and Visual Culture at the Baltic Film, Media, Arts and Communication School of Tallinn University, Estonia and a post-doctoral researcher at Aarhus University, Denmark. She is the author of the forthcoming “Selfies, why we love (and hate) them”, as well as “Body and Soul on the Internet – making sense of social media” (in Estonian). Tiidenberg is a a long time member of the Association of Internet Researcher’s Ethics Committee, a founding member of the Estonian Young Academy of Sciences, second time board member of the Estonian Sociology Association. She is currently writing and publishing on selfie culture, digital research ethics and visual research methods. Her research interests include visual self-presentation, sexuality, and normative ideologies as mediated through social media practices. More info at: kkatot.tumblr.com 

24.08. 9:30-10:30

In this lecture I will talk about how modern machine translation works, what are the requirements in order to create a new machine translation system and what kinds of mistakes such systems make. I will also talk about the motivation of using machine translation and why despite the mistakes it is still broadly usable for a number of purposes.

Mark Fishel is the head of the chair of natural language processing at the Institute of Computer Science, University of Tartu. His research includes machine translation and low-resource machine learning for natural language processing; his group also collaborates with industrial partners and has developed the online demo translation system www.neurotolge.ee.

25.08. 9:30-10:30

This is an introductory lecture to the workshop “Distant Reading by Stylometry”. The lecture will show the main tenets and methods of the field, together with examples of research in authorship attribution and distant reading.

Jan Rybicki is Assistant Professor at the Jagiellonian University in Kraków, Poland. He has written extensively on the application of quantitative methods in the study of literature, tracing the stylometric signals of authors, translators, genres and genders in literary texts in several languages. Together with Maciej Eder and Mike Kestemont, he is a co-author of the “stylo” package for R, which has become a well-known tool of stylometric analysis. He is also an active literary translator; he has translated some 30 novels from English to Polish by such authors as John le Carre, Kazuo Ishiguro or William Golding.