Place all lexical features as a lexicon in the folder (so that the process can be parallelized)

[xehu]

Currently, several of our features that use lexicons defines the lexicons inline.

For example, hedge:

hedge_words = ["sort of", "kind of", "I guess", "I think", "a little", "maybe", "possibly", "probably"]

First person pronouns:

first_pronouns = ["i",'me','mine','myself','my','we','our','ours','ourselves','lets']

For stylistic purposes, it will be better (and cleaner) to generate a file for these in the lexicons/ directory, and read it in each time. This way, we do not have to edit each python file in case the lexicon changes.

---

Files that will need to be changed include:

• Hedge.py (Hedge Words)
• Info_exchange_zscore.py (First-person pronouns)
• other_LIWC_features.py (NTRI, First-person pronouns)
• Positivity_zscore.py (Positive Affect, English Stopwords)
• Readability.py (Dale-Chall Readability Words)
• Word_mimicry.py (Function Words)

[xehu]

Also, as we update each lexicon in the list, make sure to update https://github.com/Watts-Lab/team-process-map/blob/main/feature_engine/features/lexicons/readme_lexicons.md as well! This way we can keep track of the description for each one.