Knowledge Engineering

Documentation status: guide — see Maturity and evidence.

Knowledge engineering turns domain understanding into concepts, roles, relations, constraints, rules, and contexts precise enough to validate and execute.

Fast paths

• Hypergraph
• Neuro-symbolic guide
• LLM reference

Workflow

1. identify business vocabulary;
2. distinguish concepts, instances, and values;
3. name relation roles;
4. make cardinality, constraints, and temporality explicit;
5. separate observed facts from derived facts;
6. validate the ontology or rules;
7. operationalize into logiCells models;
8. test with business cases and expected conclusions.

AI assistance

An LLM may propose terms, relations, rules, test cases, or mappings. These outputs remain candidates until human review and tool/Runtime validation.

Output

The result is not merely a diagram: it should provide stable identities and semantic commitments usable by the application.