This repository contains a collection of Knowledge Organization Systems (KOS) and related data encoded in JSKOS data format. The collection is part of project coli-conc.
Each KOS data is located in a subdirectory $ID (lowercase short name or acronym) with either one or more of the following JSKOS data files:
$ID-scheme.json$ID-concepts.ndjson
The files should be generated automatically via a Makefile. Its sources such as CSV files should be included as well. Each directory should further contain a brief README.md giving a title and links to sources and additional resources such as instructions how to update the data.
Very large vocabularies are not fully committed to the repository.
Script status.sh lists all directories and their relevant files (or selected directories passed as arguments).
If jskos-metrics is installed, vocabulary metrics can be calculated and stored in $ID-metrics.ndjson:
JSKOS_METRICS=jskos-metrics status.sh $ID
Alternatively run ./metrics.sh to update all metrics files. Calling make metrics will do so and collect the result in file metrics.json. This combined file is used for vizualization at https://observablehq.com/@nichtich/jskos-metrics.
To re-generate jskos for a given vocabulary, run:
make -sBC $ID
After having collected, cleaned and converted JSKOS data, see https://github.com/gbv/cocoda-services#usage.
Scripts are only testet on Linux. Specific requirements to generate JSKOS files depend on the source format. In any case, the following should be installed:
- make
- jq
- node and some npm modules (run
npm install) - Perl
And for some vocabularies:
- skos2jskos
- Python 3
For the usage of python you have to install a virtual environment:
- sudo apt install python3-venv
- make python
- cocoda-mappings - repository with mapping data
All data in this repository can be used freely as public domain (Creative Commons Zero v1.0 Universal)