Skip to content

process

paco xander nathan edited this page Jul 12, 2026 · 5 revisions

Ontology Process

The process employed to define the computable semantics for this domain uses the following steps:

  1. Articulate a bwyd:CompetencyQuestion to define the need for any custom term, both here in the wiki and in the domain.ttl file.
  2. Research the possible existing controlled vocabularies from which the new term can inherit.
  3. Define the term, along with its preferred title/label in both English and Français, plus its description.
  4. Link the term bidirectionally with the competency questions.
  5. Add dc:identifier references to both FoodOn and Wikidata.
  6. Publish the new definition in the domain.ttl file.
  7. Specify a SHACL shape in the shapes.ttl file to constrain the term and its usage.
  8. Populate the knowledge graph from linked data using the ontology defined above.
  9. Perform validations of the computable semantics using RDFS, SHACL, etc.
  10. Run queries, algorithmic traversals, and GraphRAG questions to substantiate the competency questions.
  11. Does anything appear to be incorrectly described, not linked sufficiently, or over-determined?

Definitions used here draw from the following primary sources:

  • Wikidata as a multi-cultural upper ontology for general definitions.
  • FoodOn as a "farm to fork" ontology about food processing and food materials, although it tends to be biased food the processing industry and research in agriculture and life science, less about usage for culinary professionals.
  • LMRI as a domain ontology for describing learning resources.
  • BBO as a BPMN-based ontology for process description.

Learning is the main activity being modeled, since learning and re-learning culinary work is the primary intent for the Bwyd language. If chefs know a process by heart, they will simply execute that. However, there are many reasons why processes must be adapted: seasonal variation for ingredients, economic constraints, food allergies for a client, and so on, There are also many reasons why a given chef or food production team will use highly specific terms (e.g., the kinds of salt) which can have dramatic effects in outcomes. The learning aspects here are especially oriented toward support for effective note-taking.

Overall the process used here is based on:

The Ontology Pipeline®
Jessica Talisman
Technics Publications (2026)

Clone this wiki locally