Karma is a data integration framework which provide - easy import from diverse sources (databases, CSV, excel files, etc...) - easy data transform (python scripts, regex and more) - easy model/mapping using preload ontologies or new ones.
For more info about Karma, check the developers repository https://github.com/usc-isi-i2/Web-Karma
The planned use of Karma on CSIRO is to help on vocabulary extraction through SKOS modeling.
This repository contains a simple script to download the main ontologies involved in this process and a Dockerfile to help the service deployment.
By running the script preloaded_ontologies_download.sh, some high level ontologies such as SKOS, PROV-O,Dublin Core, FOAF, QUDT and OP will be downloaded and a copy will be stored in the preloaded_ontologies directory. The Docker file use that files to configure Karma to load them at start.
The Dockerfile describe an Ubuntu instance of Karma running over Jetty in the port 8080.
The Karma install, compilation and service are made by Java and Maven.
See also the documentation for the karma registry pipeline