As information governance, semantics and knowledge graphs become increasingly mainstream topics, terms like ontologies and taxonomies are also becoming more mainstream. With this, we often hear questions such as: What are ontologies, How do they differ from taxonomies? How do you use them together? What value does each provide? What role do they play in a knowledge graph architecture?

Understanding the distinctions between ontologies and taxonomies is important in making decisions about metadata management which, ultimately, affects everyone who deals with enterprise data.

Ontologies describe classes of things such as materials, products, organizations or people by explicitly defining the kind of properties they may have. For example:

These properties describe objects in the world, what data we may or must collect about them and what the data should look like.

Taxonomies, on the other hand, organize data values using hierarchical relationships. For example, website topics may be organized hierarchically to support search and navigation. Specific materials could be organized into hierarchies using different organizational principles ranging from the ingredients that make up the materials to physical or mechanical properties of the materials, where the materials are found or how they are used.