What are the Types of Ontologies?
Ontologies range from simple word lists to rich logical theories. Understanding the spectrum helps you choose the right level of formality for your project.
Lightweight vs Heavyweight
Ontologies are often placed on a spectrum of expressiveness:
- Lightweight ontologies capture basic hierarchies and associations without complex logical axioms. They are easier to build and maintain but offer limited reasoning.
- Heavyweight ontologies add rich axioms — cardinality constraints, property restrictions, disjointness, equivalence — enabling automated classification and consistency checking.
Taxonomies
A taxonomy is a hierarchical classification — a tree of “is-a” relationships. Think of the Linnaean biological classification (Kingdom → Phylum → Class → … → Species) or a website’s category navigation.
Taxonomies are the simplest form of ontology. They organize concepts into parent/child relationships but do not define properties or constraints.
Thesauri & Controlled Vocabularies
A step beyond taxonomies, thesauri add relationships like “broader term”, “narrower term”, and “related term”. They also record preferred labels, alternative labels, and scope notes. SKOS (Simple Knowledge Organization System) is the W3C standard for publishing thesauri on the web.
Examples include the Library of Congress Subject Headings (LCSH), the Getty Art & Architecture Thesaurus (AAT), and UNESCO Thesaurus.
Formal Ontologies (OWL)
Formal ontologies use the Web Ontology Language (OWL) to express rich logical axioms. OWL is grounded in Description Logics and supports class restrictions, cardinality constraints, property characteristics (transitive, symmetric, functional), and equivalence/disjointness declarations.
This is the level of formality that OntoKit targets — editing and curating OWL ontologies with full expressiveness.
Domain vs Upper-Level Ontologies
Domain Ontologies
Focused on a specific subject area: biomedicine, finance, geography, etc. They define the classes and properties relevant to that domain. Examples: Gene Ontology, FIBO (Financial Industry Business Ontology), GeoNames.
Upper-Level Ontologies
Provide very general categories (object, event, quality, role) that span all domains. Domain ontologies can align to an upper ontology for cross-domain interoperability. Examples: BFO (Basic Formal Ontology), DOLCE, SUMO.
Comparison
| Type | Expressiveness | Reasoning | Example |
|---|---|---|---|
| Taxonomy | Low | Subsumption only | Linnaean classification |
| Thesaurus | Low–Medium | Broader/narrower traversal | Getty AAT, LCSH |
| OWL Ontology | High | Full DL reasoning | Gene Ontology, SNOMED CT |
| Upper Ontology | High | Cross-domain alignment | BFO, DOLCE, SUMO |