Out of Taxonomies so you’re able to Ontologies: Tailored and you may Semantic Matchmaking

Out of Taxonomies so you’re able to Ontologies: Tailored and you may Semantic Matchmaking

About example more than, the hookup next number of relationships try customized, because they are appointed from the ontologist for the sorts of circumstances

At that year’s Taxonomy Boot camp meeting, I was enjoy to provide with the committee offering 5-moment “Pecha Kucha” super discussions, wherein this year’s theme are ontology. Just as you’ll find various other understandings and you may usages off “taxonomy,” so are there other understandings and you can usages regarding “ontology.” You might arrived at if off more angles. For people who arrived at ontologies about contact with taxonomies and the field of guidance management, up coming, very just, a keen ontology is a very cutting-edge version of taxonomy who has wealthier pointers.

Within my short-term demonstration, “Away from Accidental Taxonomist so you’re able to Accidental Ontologist,” We summarized the difference anywhere between taxonomies and ontologies the following:

  1. Relationships: Taxonomies has actually hierarchical and regularly an easy “related name” associative, but ontologies has semantic relationship, which are individualized-authored.
  2. Label Properties: Taxonomies generally lack term qualities, however, ontologies would.
  3. Term Kinds: Taxonomies fundamentally don’t have categories getting terms and conditions, if you don’t envision points as kinds, however, ontologies do.
  4. Guidelines/Standards: Taxonomies is always to stick to the ANSI/NISO Z (2005) or ISO 25964, while ontologies are required to check out the web based Ontology Vocabulary (OWL) assistance and employ the Capital Description Structure (RDF).
  5. Purposes: Taxonomies support indexing/tagging, categorization, and/or group away from posts, and as a result recommendations findability and you may recovery. An important reason for a keen ontology is always to determine a website of real information, and you may help of indexing/marking, categorization, class, findability, and you may recovery will likely be second.
  6. Tools: Some software supporting the manufacture of merely taxonomies, specific software is to own ontologies, and some application can do both very well. As well, particular taxonomy/thesaurus software can assistance most, if not all, options that come with ontologies.

If you possibly could modify the relationships, performs this alter a great taxonomy into the good ontology? Zero. Designed relationships are just that function out-of an enthusiastic ontology, even in the event even the essential feature. Inside my on the internet course on the taxonomies, even in the event I don’t teach how to create ontologies, I really do provide a lesson with the customized/semantic dating. It is often liked by manage a more advanced taxonomy instead fundamentally meeting all the criteria out-of an enthusiastic ontology.

The latest generic RT relationships, although not, is actually singularly bidirectional

Also, a customized dating is almost certainly not totally semantic. The fresh new relationships also are “semantic” as they contain particular definition. (Semantic function “have definition.”) You can easily modify matchmaking when you’re nevertheless perhaps not which makes them totally semantic. Your age the standard matchmaking for your form of application and you can listeners. Particularly, you can rename greater title (BT)/narrower identity (NT) because “ parent/kid ,” or rename Related Title just like the “ see along with .” In case your taxonomy/thesaurus application is more sophisticated, it can allow you to establish any number of tailored relationship, and thus you could add significantly more subtleties out of meaning.

An essential component from it is semantic relationships sure-enough when you look at the ontologies ‘s the capacity to create directional dating which might be type of inside the for each recommendations, which have reciprocity. Most of these semantic relationship might possibly be variations off “ related term ” (RT), unlike alternatives of your own hierarchical relationship. If you merely tailored they of the renaming it, it would have to be a comparable in recommendations, like features “ has actually lover .” To create an effective semantic relationships few, eg Boy ( makes ) and you will IND ( produced by the ), you need a tool that supporting ontological matchmaking and not “customized” matchmaking.

In case your unit supports designed relationship but not the capacity to would line of sets out-of directional relationship that are associative in the place of hierarchical, the outcome cans nevertheless be very beneficial. You may have a good “close ontology” if you don’t a purely discussed ontology. Such as, you might rename the fresh new just one “ relevant term ” (RT) as the “ Manufacturer-Produc t” having an acronym including Man-Expert (Borrowing from the bank to help you Alice Redmond-Neal out-of Availableness Inple). Therefore, the connection is the identical in a choice of advice:

Αφήστε μια απάντηση