Agenda

Led by associate professor Eetu Mäkelä, the HSCI research group seeks to figure out the technological, processual and theoretical underpinnings of successful computational research in the humanities and social sciences. In doing so, we develop technical tools, algorithms and workflows, but also study ways to better structure the research process overall. We also work on the theoretical level, for example evaluating the epistemic soundness of different ways of integrating qualitative and quantitative modes of research.

Our primary approach in doing this is to partner simultaneously with multiple projects in the humanities and social sciences. Through working with projects that operate in different subfields, but still have commonalities in either data or analysis needs, we are able to see past individual scenarios. This way, we are able to identify and target broader classes of issues, both in tool building as well as the overall process of computational research itself.

Projects, research groups and institutions we are working, or have worked with include HSSH, FLOPO, FILTER, COMHIS, HELDIG, NDHL, STRATAS, CofK, RROL and Humanities+Design in addition to many smaller collaborations.

Current Interests

• The inherent problems in non-standard big data for research: bias, confounders, the gap between what is in the data and what is of interest, and the need for cleanup. For an initial report, see Wrangling with non-standard data, or e.g. this presentation
• How to best organize interdisciplinary research. For initial reports, see An Approach for Agile Interdisciplinary Digital Humanities Research – a Case Study in Journalism, and Interdisciplinary Collaboration in Studying Newspaper Materiality.
• How to develop technical tools and workflows 1) efficiently and 2) to cater to as broad demands as possible. Results not yet uniformly reported, but see e.g. A Workflow for Integrating Close Reading and Automated Text Annotation, this presentation and this listing.
• Identifying epistemologically sound ways of combining quantitative computational analyses and qualitative interpretation. Not uniformly reported, but see e.g. Topic Modeling and Text Analysis for Qualitative Policy Research for initial thoughts on a particular method in a particular context.