Liquid-Legal-Institute/Legal-Ontologies
-
What it is — A curated list of selected resources, methods, and tools for legal ontologies, data schemes, and knowledge graphs. It is an index rather than a single ontology: its sections point to use cases, ontologies, datasets, modeling tools, research groups, and tutorials.
-
Key concepts — The documented ontology-learning tasks include domain terminology extraction, concept discovery, concept-hierarchy derivation and extension, learning non-taxonomic relations, rule discovery, ontology population, and frame/event detection. The catalog also shows the breadth of legal modeling targets through resources for legislation, legal documents and rules, contracts, copyright, transactions, legal services, and parliamentary activity.
-
How you’d use it — Use the README as a discovery map: follow its links to candidate legal ontologies and standards, supporting datasets or annotation guidelines, and tutorials. For modeling and development it names Protégé, Fluent Editor, OntoStudio, and OWLGrEd; listed interchange resources include Akoma Ntoso, LegalRuleML, Legal Document ML, and UBL, which the README describes as XML templates for potentially legally binding business documents.
-
LLM angle — none stated
-
Pitfalls & lessons — The list cautions that GoodRelations is not strictly legal, although it may help model transactions. Maintenance is category-driven: contributors are asked to submit one suggestion per pull request, add links at the bottom of the relevant category, use informative titles, and explain why an addition is useful; new categories and categorization improvements are welcome.
-
Verdict — A concise starting index for finding legal ontology resources and tooling, not a self-contained legal domain model or implementation guide.
Sources consulted
README.mdcontributing.md