The core Innovonto-Knowledge Graph Server: Contains a Fuseki-Based Triplestore and a react-based frontend
At the moment, the project uses a submodule to include the Innovonto-Ontology, located at https://github.com/FUB-HCC/Innovonto-Ontology
Read more about submodules here
If you want to get build a deployment artifact, check out the project using:
git clone --recurse-submodules https://github.com/FUB-HCC/Innovonto-Core.git
TODO add dev commands here
Step 1: Build the different artifacts (TODO move into docker-build)
1) Entity Server: yarn build
2) Ontology: make
3) idea-map-backend
Step 2: run docker-compose build
Step 3: push
Fuseki is a triple-store combined with a SPARQL-server it provides the backend for the basic entity-server displays and a public SPARQL-endpoint to access the idea knowledge graph. Read more about fuseki here.
In order to build an idea-map, we use a python-based pipeline to build vector-embeddings from the ideas, cluster these and then reduce the embeddings to a 2-Dimensional representation. These steps are done by the python app located at /services/idea-map-backend
see https://github.com/FUB-HCC/Innovonto-Ontology for docs.