29 test category mapping techniques

[cmbrennan002]

---

Description

This is a draft PR, created to hold work over between sprints. So far, the code:

• Expands the keyword search, then uses keywords to extract all sentences with a particular subcategory
• Create phrases for matching from these sentences (manually)
• Calculates the cosine similarity between all phrases labelled as 'benefits' in the binary labelled dataset, and the phrases for each subcategory

Next steps:

• Iterate through phrases to create cosine similarity for the full dataset
• Use the max score, and average score, to predict final category
• Increase range of categories
• Improve efficiency of scores
• Shift to using benefit classifier output (rather than currently the labelled input)

Fixes # (issue)

Instructions for Reviewer

In order to test the code in this PR you need to ...

Please pay special attention to ...

Checklist:

• I have refactored my code out from notebooks/
• I have checked the code runs
• I have tested the code
• I have run pre-commit and addressed any issues not automatically fixed
• I have merged any new changes from dev
• I have documented the code
  • Major functions have docstrings
  • Appropriate information has been added to READMEs
• I have explained this PR above
• I have requested a code review