https://github.com/adamspannbauer/nlhp

Random NLP tasks with Harry Potter text
https://github.com/adamspannbauer/nlhp

Last synced: about 2 months ago
JSON representation

Random NLP tasks with Harry Potter text

• Host: GitHub
• URL: https://github.com/adamspannbauer/nlhp
• Owner: AdamSpannbauer
• License: mit
• Created: 2020-01-20T13:56:24.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2020-12-31T14:30:36.000Z (over 5 years ago)
• Last Synced: 2025-01-13T05:28:45.229Z (over 1 year ago)
• Language: Jupyter Notebook
• Size: 31.7 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# nlHP

Random NLP tasks with Harry Potter text.  Probably written in python... might use R for more nostalgia and for Harry Pot-R pun potential.

Data gathered from [bradleyboehmke's `harrypotter` R package on GitHub](https://github.com/bradleyboehmke/harrypotter) (tbh, not sure legality of the whole thing, so the data directory is excluded from the repo to avoid redistributing any potentially legally sketchy stuff).

## House Wordclouds

#### Workflow

* Parse books into sentences with `spacy`

* Filter sentences to where each of the houses were mentioned

* Create a wordcloud with each house's sentences using their Pottermore crest as the mask/coloring

#### Output