Ontologies 101


When was the last time your computer discovered something you didn’t explicitly tell it about? If it’s been a while, maybe you need an ontology. Ontologies – formal specifications of the objects in a domain and the relationships between them – let you make models of things you know about, and then reason over them to discover new relationships. Ontologies are powerful, dynamic, and they let you organize your data much closer to the way humans think than a database can. They’re also one of the technologies at the heart of the emerging Semantic Web (Web 3.0).

In this talk, I’ll give a brief introduction to ontologies, how they’re used as knowledge modeling tools, and the kinds of reasoning you can do with them. We’ll touch on some of the more interesting things that come up when you use ontology-based systems, like the need for confidence and lineage. We’ll cover the differences between ontologies and databases, and lay the famous “I could do that with my database!” argument to rest. I’ll wrap up with a brief intro to the Semantic Web, the place you’re most likely to run into an ontology. You’ll come away from this talk (finally) understanding what semantics are, and with enough information to start building your own ontologies.

Sharon Stern


Sharon Stern has a degree in Comparative Literature, and an interest in linguistics and computing. She architects ontologies at Thetus, a leading provider of semantic modeling software