Data model for Knowledge Graph

[AlexMikhalev]

Atomic Data Server seems to be a perfect fit for storing Knowledge Graphs, like https://www.nlm.nih.gov/research/umls/knowledge_sources/metathesaurus/index.html or https://github.com/commonsense/conceptnet5 Knowledge Graphs structure is normally: Ontology (Types of entities)->Taxonomy (How they are connected together)->Term (Word). For example in conceptnet node baboon: API http://localhost:8084/c/en/baboon

"@id": "[/c/en/baboon](http://localhost:8084/c/en/baboon)",
  "edges": [
    {
      "@id": "[/a/[/r/IsA/,/c/en/mandrill/n/wn/animal/,/c/en/baboon/n/wn/animal/]](http://localhost:8084/a/[/r/IsA/,/c/en/mandrill/n/wn/animal/,/c/en/baboon/n/wn/animal/])",
      "@type": "Edge",
      "dataset": "[/d/wordnet/3.1](http://localhost:8084/d/wordnet/3.1)",
      "end": {
        "@id": "[/c/en/baboon/n/wn/animal](http://localhost:8084/c/en/baboon/n/wn/animal)",
        "@type": "Node",
        "label": "baboon",
        "language": "en",
        "sense_label": "n, animal",
        "term": "[/c/en/baboon](http://localhost:8084/c/en/baboon)"
      },

Node with label baboon and language English is term have a relation isA (ontology) animal (Taxonomy) and have a relation Synonym (Taxonomy). The ask is to create a data model which allows to capture terms and their relationship, full list for conceptnet is here. Firstly, I am interested in isA and Synonym relationships. If conceptnet is too difficult (or too large) we can start with wordnet.