This repository contains supplementary materials for Chapot, Laura. "Computational Resonance" in Canadian Review of Comparative Literature.
To consult the array of topic models used for analyses click on array-of-topic-models above.
The directory topic-model-results contains the original topic model outputs used for developing the analyses.
I generated a number of topic models of the corpus varying the number of topics (from 5 to 10) and running each topic model twice. I used two topic model outputs: lists of topics with their most frequent words and lists of distributions of topics over each document in the corpus. For example, topic-model-results/mann-topwords-5 is a list of topics with their most frequent words with a parameter of 5 topics. And topic-model-results/mann-topicsperdoc-5 is the distribution of those topics over each document in the corpus.
The directory mann-novellas-corpus contains the corpus of Thomas Mann's early novellas used for generating the topic models.
The texts that compose this corpus include the collection Der kleine Herr Friedemann (1898) — which comprises the stories “Der kleine Herr Friedemann” (1896), “Der Tod” (1897), “Der Wille zum Glück” (1896), “Enttaüschung” (1896), “Der Bajazzo” (1897) and “Tobias Mindernickel” (1898) — and the novellas Gladius Dei (1902), Tristan (1903), Tonio Kröger (1903), Schwere Stunde (1905) and Der Tod in Venedig (1912).
N.B. This corpus is not lemmatized.
The collection of novellas contained in Der kleine Herr Friedemann (1898) has been preserved as a collection, it has not been split into the individual constitutive stories.
I will be adding Notebooks to generate your own topic models of the corpus.