Automated content analysis

31 mars et 1er avril 2017

Jonathan Miaz, Coordinateur du PDSPO, Université de Genève

Bruno Wueest, University of Zurich

Automated content analysis (ACA) is one of the key fields of methodological innovation in the social sciences and is on the verge of becoming a standard tool for political scientists. In this two day workshop, I will first present the basics needed to get started with ACA. More precisely, I will provide the IT required to conduct ACA in the statistical software package R and discuss the all so important concepts of machine learning and validation. Second, I will give an overview over the most frequently used approaches, which are dictionary based coding, supervised text classification and unsupervised text clustering. The goal is to show how you as social scienteists can profit from the tools ACA has to offer. Hence, for every approach, we will look at its methodology, some examples from the literature and selected implementations in R. The workshop will be designed in an interactive and application-oriented way. This also means that you are encouraged to bring your own text sources and research questions so we can discuss the feasibility of different tools for your research.

Université de Lausanne, Geopolis salle 2208

20