Jupyter Notebooks Relating to Open Context (https://opencontext.org). These notebooks will help people get started using Open Context in "open science" and reproducible research workfolows. Please feel free to fork and adapt this repo for your own needs.
This is very much a work in progress as I'm still a novice with Jupyter Notebooks and some of the Python data visualization and scientific computing libraries. But I'll try to periodically update this for use
- Notebook to download a search/query result records and save locally as JSON and CSV. Caching data locally will be very useful because big data dumps from the API take lots of time.
- Notebook to provide citation information for search/query results. If you want to use Open Context for research it is important to cite the data creators. I intend to make this easy with some code that requests full citation information for the different projects that may be represented in a search/query result set.
- Notebook to generate scatter plots of the most commonly co-occuring numeric attributes. This will be useful to get a general sense about variability within a search/query result set.
- Some machine-learning experiments. Because if the world will inevitably have a SkyNet AI, that AI should have esoteric obsessions with ancient stuff.
Most of the recent development for these notebooks took place using an Ubuntu 18.04 subsystem on Linux. I found it easier to install various Python packages in Linux world rather than on Windows directly. However to start Jupyter Notebooks in the Linux subsystem, I needed a special invocation as follows:
jupyter notebook --ip=127.0.0.1 --port=8888