This is a repository to interpret submission data and serve it up as tasty tasty jam JSON for visualisation and further analysis.
The hardest problem in computer science is naming things. The Lab team avoids this unnecessary obstacle by naming projects after what we have for breakfast. The bagel project is a series of discovery/research sprints, exploring emerging tools for policy consultation analysis. This is one of the repos that was created in our analysis.
It has a few parts, and some more READMEs:
- A subset of submissions from the Zero Carbon Bill and a Te Reo Māori corpus - more details in the READMEs for those folders
- Amazon Comprehend's topic modelling, key phrase detection, and sentiment analysis of these
- Jupyter notebooks to play with the data - these are interactive at Binder
- The JSON files in
docs
are served with GitHub Pages export_lda_viz.py
exports the LDA analysis to a visualisation indocs/lda/index.html
To run notebooks locally:
# Install correct python version (you'll need pyenv)
pyenv < .python-version
# install dependencies
pip install -r requirements.txt
# spin up the notebook
jupyter notebook