From the top level folder run: python ./power_agency_gender/power_score_gender.py

Code available in booknlp_output/gender_agent_booknlp.ipynb

Note that the Tereus/Philomela output is manually edited to be more accurate. The code can be run on other stories, but power/agency scores will be less accurate per character.

• topic-modeling/lda-train.py: training file for LDA topic model. From the top-level folder, run python ./topic-modeling/lda-train.py ovid custom. This will run topic modeling on our dataset with a custom stopword list.
• topic-modeling/lda-prep.ipynb: segment our dataset into documents for training.
• topic-modeling/lda-analysis.ipynb: analyses on LDA results

• booknlp_output/entire_wordcloud.ipynb: analysis of word clouds on the entire text