A working example

The AssociationAI knowledge graph

This page is a small knowledge graph of this site: 26 articles, 15 tutorials, 13 hands-on tools, and 10 topics, connected by 154 relationships. It is built the way Neo4j's documentation teaches graphs: nodes are the things (each with a label like Article and properties like a title), relationships are the named, directed connections between them (like [:COVERS]). Drag nodes, scroll to zoom, and click anything to inspect it. If you read Association Ontology, this is the same idea, running on the site itself.

How to read this graph

Node labels

(:Article) Explains a topic.
(:Tutorial) Teaches a task, step by step.
(:Tool) A hands-on demo page you can try.
(:Topic) A subject several posts cover.

Relationship types

(t:Tutorial)-[:INCLUDES_DEMO]->(tool:Tool) The tutorial links to this hands-on demo page.
(c)-[:COVERS]->(tp:Topic) The article or tutorial covers this topic (from its tags).
(a)-[:REFERENCES]->(b) The post links to another post on the site.
(tool:Tool)-[:SUPPORTS]->(tp:Topic) The tool supports this topic, through the tutorial that includes it.

In Neo4j's Cypher language you would ask for part of this graph with a pattern like MATCH (t:Tutorial)-[:INCLUDES_DEMO]->(tool:Tool) RETURN t, tool. The buttons below run traversals like that and highlight what matches.

Try a traversal:

The dataset is generated from the site's own content by scripts/build-content-graph.mjs, so the graph stays in sync as posts are added. Relationships come only from what the content itself says: tutorial pages linking to demos, posts linking to other posts, and tags. Graph concepts follow Neo4j's graph database documentation and the Cypher introduction.