Writing brief. This monograph has been assigned but the manuscript is not yet written or approved. What follows is the commissioning brief: its question, outline, evidence instructions, and review requirements.
Central question
What follows from a formal definition?
Reader promise
The reader will be able to read a small set of axioms, predict what a reasoner will conclude, say which axiom each conclusion depends on, and recognize when a conclusion is an artifact of a modeling choice.
Required development
- Introduce concepts, roles, and individuals; subsumption as "every instance of A is an instance of B"; consistency as the absence of contradiction; and decidability as the guarantee that a reasoner will finish. Explain the TBox and ABox division and how it mirrors the class and instance distinction.
- Provide a small worked inference. State three or four axioms in a named description logic (for example the logic underlying the OWL 2 EL profile), declare the individuals, and derive one subsumption and one classification by hand. Then show one plausible conclusion that does not follow and say which axiom is missing.
- Connect to ontology reasoning. Explain what a reasoner does (classification, consistency checking, realization), which OWL 2 profiles correspond to which logics, and why the profile determines what can be expressed and how fast reasoning runs.
- Use a synthetic nuclear medicine example with local identifiers: define "diagnostic radiopharmaceutical" and "therapeutic radiopharmaceutical" by axioms, add a synthetic agent, and show whether and why it is classified under either.
Evidence and review
Evidence to obtain: the Description Logic Handbook and the 2017 introduction for formal definitions and notation; the OWL 2 Direct Semantics for the mapping from OWL constructs to description logic; the OWL 2 Profiles document for the EL, QL, and RL restrictions; and SNOMED International's statement of which logic its concept model uses, cited to the current release. Every worked inference must be reproducible by a named reasoner, with the reasoner and version recorded.
Review concern: a reasoner's conclusion may be presented as a discovery about the domain. The manuscript must show, for every derived statement, the axioms it depends on, and state that an inference is only as true as its premises and its modeling assumptions.
Planned internal links
Description logic, Reasoning, OWL, Ontology for classes and instances, Knowledge graphs for traversal versus entailment, and Semantic web standards for the encoding. Link at the point of use.
Review requirements
Source check against the specifications. Technical review by someone who has built and classified an OWL ontology, re-running the example. Editorial check that identifiers are visibly local.
Review gate for this monograph
Show the axioms and modeling assumptions.
The assignment exists. The manuscript has not been written or approved.