AI series · Part 1: GenAI and Knowledge Modeling
Ontology: The Blueprint Behind Enterprise Knowledge
A taxonomy says what is a type of what. An ontology also says how things relate and which rules they follow. Here's the difference, with Schema.org as a real-world example.
A taxonomy only says "is a type of". An ontology goes further: it defines how things relate and the rules they follow. It is the blueprint of your domain. It describes what can exist, but it holds no actual data.
- An Employee works in a Department
- An Employee reports to a Manager
- A Manager is also an Employee
A real-world example: Schema.org
Google uses Schema.org to understand websites. It's an ontology that says a Person worksFor an Organization and a Product has a Price. When a website marks up its pages this way, Google can show prices and star ratings directly in search results.
The ontology in our HR assistant
Our reference build defines these classes and relationships:
| Kind | Defined in the ontology |
|---|---|
| Classes | Employee, Manager, Department, Policy |
| Relationships | worksIn, reportsTo, governedBy |
Notice there's no "Rahul" or "Sales" here. Those are data, and data lives in the knowledge graph, which we cover next.
An ontology is the formal schema of a domain: its classes, relationships and rules. The knowledge graph is the real data that follows it.
- The short version
Next in the GenAI series: Knowledge Graphs: Answering the Questions Vector Search Can't
Start from the top: GenAI Quick Answers and Series Guide
We deploy these patterns across production engagements in fintech, healthcare, and eCommerce. Talk to a senior engineer about how they apply to your platform.