An ontology specifies the categories of entities, the relationships between them, and the constraints on those relationships. In litigation it defines types such as party, witness, provider, document, event, claim, and defense, along with relationships such as testified about, supports, and contradicts.
It determines what questions the system can answer structurally.
Alternative Names:
Domain Ontology, Knowledge Schema
Why it Matters?
The ontology is where domain expertise enters a legal AI system, and it is the difference between a generic graph and a useful one. Encoding that a fact can be disputed, that a witness is the source of an assertion, and that a document supports or contradicts a claim is what allows structural questions to be answered. Systems built without a litigation-specific ontology produce connections that are technically correct and practically useless.
Frequently Confused with
Related terms
Frequently asked questions
What does a litigation ontology define?
Why does the ontology matter more than the graph technology?
