Try it: build your first association knowledge graph

This is the working version of our knowledge graph tutorial, running right here in your browser. The finished Riverbend graph is already loaded below: click any dot to walk its connections. Then build your own from a CSV. Nothing you paste ever leaves this page.

Click any dot to walk the graph. Every connection becomes a button.

The finished Riverbend Trade Association graph: 18 nodes, 20 edges. Fictional data, built from the sample CSV.

Build from a CSV

The sample export is loaded below. Paste your own member export over it, or pick a file from your computer, then map the columns. The builder runs entirely in your browser: no upload, no server, no account.

Your CSV (headers in the first row)

Name the connections

Tick the relationships your columns can support. Each one becomes a named edge between the person in a row and the thing in that row's column. Use verb phrases, the way you would say the sentence out loud.

people work for companies people belong to chapters (edge: member of) people serve on committees people hold certifications

What you just built

Entities (the named things), attributes (facts that stay with their person), and relationships (named connections between things). That is the whole model: the W3C's RDF 1.1 Concepts describes graph data the same way, as subject-predicate-object triples. The demo skipped the attribute columns on purpose; they ride along with their person rather than becoming dots of their own.

Read the full tutorial for the mapping decisions behind each step.

Download the starter kit

The kit holds the sample CSV and a one-page mapping worksheet: the entity types, the attribute-vs-relationship test, and blank lines for your own columns. Free with your email.