On October 2020 we hosted an online workshop in collaboration with scholars from the Centre for Digital Social Research at Umeå University (DIGSUM) and the Welfare State Analytics (Westac) group at Umeå University to discuss issues and advances in performing text analysis on political, traditional media, and social media data.

News pappers on the table

We assembled an interdisciplinary team to share their experience using various types of textual data sources, ranging from official government documents to online tweets. In addition, we had sessions focusing on the massive newspaper corpora our team is assembling in conjunction with KBLab at the National Library of Sweden. We discussed the advantages and obstacles of these data sources and participants shared the latest research results derived from these corpora.

Mining for Meaning (IAS at LiU)

  • Miriam Hurtado Bodell, Måns Magnusson, & Sophie Mützel, on curating the KB data
  • Sarah Valdez & Anastasiia Menshikova, on the Swedish COVID discourse on Twitter
  • Miriam Hurtado Bodell, on measuring shared understandings using seeded topic models on KB data

DIGSUM (Umeå University)

  • Simon Lindgren, on interpretation/computation of climate denialism on Twitter
  • Johan Jarlbrink, on studying court judgements using automated text analysis, machine learning, and network analysis
  • Moa Eriksson Krutrök, on the expanded discourse of terrorism through hashtag co-occurrences
  • Sam Merrill, on combining social media and (auto)ethnographic analysis

Westac (Umeå University)

  • Johan Malmstedt & Erik Edoff, on the concept of political in Swedish newspaper data
  • Pelle Snickars, on topic models in Swedish SOU data

 We thank our funder Vetenskapsrådet for making this very fruitful event possible.