SODAS Data Discussion w/ Aske Mottelson & Hjalmar Alexander Bang Carlsen
Copenhagen Center for Social Data Science (SODAS), is pleased to announce that we are continuing with SODAS Data Discussions this fall.
SODAS aspirers to be a resource for all students and researchers at the Faculty of Social Sciences. We therefor invite researchers across the faculty to present ongoing research projects, project applications or just a loose idea that relates to the subject of social data science.
Every month two researchers will present their work. The rules are simple: short research presentations of ten minutes are followed by twenty minutes of debate. No papers will be circulated beforehand, and the presentations cannot be longer than five slides.
The first presenter is Aske Mottelson. He is Ph.d in Computer Science and Postdoc at the Department of Psychology, University of Copenhagen. Aske is interested in the intersection between human behavior (social psychology, personality psychology) and computational methods (machine learning, data science, software engineering). His recent work includes research in Affective Computing, Mobile Sensing, and Virtual Reality. The second presenter is Hjalmar Alexander Bang Carlsen, Postdoc here at SODAS.
Aske Mottelson: Investigating online political candidate tests from the 2019 Danish general election by scraping and running random simulations
Online tests that match voters with political candidates are increasingly popular ahead of elections. In this data discussion, I will present scraped data, simulation results, and speculations about the popular political tests from the 2019 Danish general election. The data suggest some candidates might have 1000x probability of being suggested by these platforms, because of the choice of algorithms to compute similarity. Hopefully, this will initiate discussions about the algorithmic issues (ethically and mathematically) of similarity scoring between questionnaire responses from voters and political candidates.
Hjalmar Alexander Bang Carlsen: Qualifying text as data: steps towards interpretatively valid and unbiased classification of textual data
Classification of text has huge potential for digital social research allowing researchers to get qualitative information on social interactions at scale. Yet very tricky methodological concerns arise when wanting to classify large amounts of text. Many of the textual dataset newly available to social scientists are heterogeneous - spanning very different populations and practices. In such a setting concerns with interpretative validity and biased measurement error become particularly acute. In this talk I will present two of the dominant strategies within text classification which tries to handle these difficulties: computationally grounded theory and supervised classification. I note the limitations in both strategies and proceed to present a set of steps which seek to ensure both interpretatively valid categories and good measurement.
The SODAS Data Discussion will take place at SODAS in building 1, 2nd floor, room 26 (1.2.26) of the CSS Campus, University of Copenhagen, from 11.00 am to 12.00 noon.
If you have questions or want to know more, please write Sophie Smitt Sindrup Grønning at firstname.lastname@example.org.