Supervisor: Gerold Schneider
We have large historical text collections, for example the English COHA corpus, the NZZ archive or Journal de Genève. What trends and stories do we find when using distance reading approaches? You will apply topic modeling, Kernel Densitiy Estimation, automatic summarisation, document classfication, keyword detection etc. to investigate a historical topic of your choice. While it is difficult to evaluate these approaches, you will need to come with clear and verifiable hypotheses.