Learning Graph JSON Schema

Hierarchical structure of the learning graph JSON format for vis-network

graph TD Root["learning-graph.json"]:::rootNode Meta["metadata"]:::metaNode Groups["groups"]:::groupNode Nodes["nodes (array)"]:::nodeNode Edges["edges (array)"]:::edgeNode Root --> Meta Root --> Groups Root --> Nodes Root --> Edges Meta --> M1["title: string"]:::fieldNode Meta --> M2["description: string"]:::fieldNode Meta --> M3["creator: string"]:::fieldNode Meta --> M4["date: ISO 8601"]:::fieldNode Meta --> M5["version: string"]:::fieldNode Groups --> G1["FOUND: {color, font, shape}"]:::fieldNode Groups --> G2["BASIC: {color, font, shape}"]:::fieldNode Groups --> G3["... (other taxonomies)"]:::fieldNode Nodes --> N1["[0]: {id, label, group}"]:::fieldNode Nodes --> N2["[1]: {id, label, group}"]:::fieldNode Nodes --> N3["... (~200 objects)"]:::fieldNode Edges --> E1["[0]: {from, to}"]:::fieldNode Edges --> E2["[1]: {from, to}"]:::fieldNode Edges --> E3["... (~600 objects)"]:::fieldNode classDef rootNode fill:#ffd700,stroke:#333,stroke-width:3px,color:#333,font-size:16px classDef metaNode fill:#4facfe,stroke:#333,stroke-width:2px,color:#fff,font-size:16px classDef groupNode fill:#43aa8b,stroke:#333,stroke-width:2px,color:#fff,font-size:16px classDef nodeNode fill:#9b59b6,stroke:#333,stroke-width:2px,color:#fff,font-size:16px classDef edgeNode fill:#f77f00,stroke:#333,stroke-width:2px,color:#fff,font-size:16px classDef fieldNode fill:#ecf0f1,stroke:#333,stroke-width:1px,color:#333,font-size:14px linkStyle default stroke:#999,stroke-width:2px

JSON Structure Overview

The learning graph JSON format is designed for vis-network visualization and contains four main sections: metadata, groups, nodes, and edges.

Main Sections

1. metadata (Dublin Core)

Descriptive information about the learning graph following Dublin Core standards.

2. groups (Visual Styling)

Defines colors, fonts, and shapes for each taxonomy category (FOUND, BASIC, ARCH, etc.).

3. nodes (~200 objects)

Array of concept objects, each with:

id: Unique integer identifier
label: Concept name (max 32 characters)
group: Taxonomy category code

4. edges (~600 objects)

Array of dependency relationships, each with:

from: Prerequisite concept ID
to: Dependent concept ID

For a 200-concept graph with average 3 dependencies per concept, expect ~600 edges.

Example JSON

{
  "metadata": {
    "title": "Introduction to Learning Graphs",
    "creator": "Claude AI",
    "date": "2025-11-17"
  },
  "groups": {
    "FOUND": {"color": "#ef5350", "font": {"color": "#fff"}},
    "BASIC": {"color": "#ff7043", "font": {"color": "#fff"}}
  },
  "nodes": [
    {"id": 1, "label": "Graph Theory", "group": "FOUND"},
    {"id": 2, "label": "Dependencies", "group": "BASIC"}
  ],
  "edges": [
    {"from": 1, "to": 2}
  ]
}