Mary Wollstonecraft and Influence - Preprocessed Data

Preprocessing steps for this topic modeling are laid out in this dataset. It contains a stopword list, additions made to the stop word list, plus steps taken to clean each text within the corpus.

Original stopword list obtained at GitHub: <a href="https://github.com/fozziethebeat/S-Space/blob/master/data/english-stop-words-large.txt">https://github.com/fozziethebeat/S-Space/blob/master/data/english-stop-words-large.txt </a>

For this project, preprocessing was a much less straightforward step than the collocate/concordance project from Fall 2017. Most of the preprocessing was done by myself [Scott Caddy] via SublimeText. Segmenting of items in the corpus was done via the TopicModelingTool settings. What I have here is a step-by-step process attempted over several weeks as I learned more about the TopicModelingTool, topic modeling, and how to interpret the results.

A large part of my preprocessing came from doing topic modeling "runs" and modying my stopword list according to results. Hopefully, this component can show how I continued to hone preprocessing, based on feedback from my peers, instructor, and reading papers on topic modeling.