Semantic Networks

A semantic network, also called a knowledge graph, represents relationships between concepts using a directed graph of labeled nodes and labeled arcs. Each node represents a concept, category, or instance; each arc connects an ordered pair of nodes and describes the relationship between them, expressed as a subject-predicate-object triple. Four arc types are especially common: is-a links, which describe class and subclass relationships and form inheritance chains, the most common type; has-a links, describing an attribute or characteristic of a node; inst links, describing an instance-of relationship; and part-of links, describing a whole/part relationship.

Inheritance is central to how semantic networks work, a child node connected to a parent by an is-a link inherits the parent's attributes automatically, which is what makes the network more than just a collection of disconnected facts.

A semantic network is a taxonomy that's been given a vocabulary for the connections themselves, instead of just "this is under that," it can say exactly how one thing relates to another, and let a machine follow that reasoning.

Exercise

The scenario: A museum is building a knowledge graph to connect its collection objects, artworks, artists, historical periods, geographic origins, materials, and techniques. Five relationships between collection objects need to be represented. For each one, identify the correct arc type.

1. "This bronze sculpture was made using the lost-wax casting technique."
2. "This specific vase is an example of Ming dynasty porcelain."
3. "Sculpture is a kind of visual art."
4. "This fresco is part of the Sistine Chapel ceiling."
5. "This painting has the attribute 'oil on canvas' as its medium."
Try It With Your Data: Polyhierarchy Analysis Pattern

Fill in your own details below; the prompt updates as you type. When it's ready, copy it into Claude or whatever AI tool you use.

What this prompt is made of (RICE breakdown)
R

Role

Casts the AI as: You're an information architect analyzing content for polyhierarchy.

I

Instructions

Names 4 specific outputs to produce, so the response comes back structured rather than a general summary.

C

Context

User types, Common user tasks and Navigation constraints tell the AI what your specific situation is, not a generic one.

E

Expected format

For each polyhierarchy item: the dual homes, why both are justified (or not), and a recommended primary placement with cross-links.

Paste your current taxonomy structure
Comma-separated list
Comma-separated list
Any limitations to work within
Assembled prompt