An ontology is a structure that defines the concepts and relations of a field. This page explains what it does in ONTOFLOW, split into data, rules, decisions, holds and access.
01 · Data
Connects data and keeps the source.
Databases, documents, files, sensors and regulations are tied into one knowledge graph. Every value keeps where it came from. Names of concepts and relations follow the term registry for each field.
See it on a public page · Korean industrial term registry →
02 · Rules
Checks against rules.
The rules that the ontology and data must follow are written in standard forms such as SHACL, and violations are found. Ontology, rules and graph export to OWL, SHACL and Turtle.
See it on a public page · Ontology checker →
03 · Decisions
Records evidence for each decision.
Each decision carries its clause, record, source and the date it applies to. You can replay the path the evidence took as a graph. Signed reports keep a number and a body hash.
See it on a public page · Workplace duty check · evidence graph →
04 · Hold
Holds back when evidence is short.
Missing values are not filled with guesses. When a value needed for a decision is missing, the answer is "cannot determine", with what is missing (the data gap) written alongside.
See it on a public page · Workplace duty check · leave the content blank to see "cannot determine" and the field to fill →
05 · Access
Guarded by access control and audit logs.
Keycloak single sign-on and role- and attribute-based access decide who sees what. Who did what, and when, is kept in a hash-chained audit log.
This runs inside your network, so there is no public page.
This page describes features that exist in ONTOFLOW. There is no customer operating record yet. Legal outputs are reference material, not legal advice.