What is Semantic Triple?
- Semantic Triple
- A semantic triple is a single fact expressed in three parts, a subject, a predicate and an object, as in the statement that an organisation provides a service. Semantic triples are the shape knowledge graphs store facts in, and structured data on a page is a way of handing an engine those triples directly rather than leaving them to be inferred from prose.
01 In practice
JSON-LD on a page is a set of triples in a different notation. An Organization node carrying a name, a service type and an area served is three statements about one subject, and each of them can be joined to statements another source makes about the same subject.
Prose expresses the same facts less reliably. A sentence saying a business has served a region since a given year asks the reading model to identify the subject, resolve which business is meant, and interpret the claim, and each of those steps can go wrong in ways a triple cannot.
Writing with triples in mind is a discipline rather than a format. Sentences shaped as subject, relationship, object are the ones a model extracts most cleanly, which is the same reason a definition that names its own subject outperforms one that assumes it from the heading.
02 Related terms
The terms next to this one.
The work behind it
Tell the models what you are.
Markup that tells models what your business is, on pages that render without JavaScript.