Guest lecture on probabilistic topic models

21.2.2020 at 13:00-14:00

407, Agora, University of Turku,

Department of Future Technologies have invited Postdoctoral researcher Måns Magnusson from the Aalto University to give a guest lecture on Using priors to capture concepts of interesting large corpora.

Large-scale probabilistic models, such as word embeddings and topic models, are becoming more popular among researchers in the social sciences and the humanities. The hope is often to understand and use large corpora in scientific inquiry. Although, the fully unsupervised approach of these models makes it difficult to draw scientific conclusions.

Måns Magnusson will present a new simple way of using priors to identify the concept of interest for researchers in the social sciences and humanities, using both probabilistic topic models and word embeddings. The approach is a straightforward and principled framework for confirmatory scientific inquiry of semantic content in large textual corpora.