Re: OWL mismatch example: Taxonomies VS. the semantic web?

Benjamin Nowack <office@e-senses.de> wrote: 

>hmm. wouldn't the only question you can ask a reasoner be:

>"which articles exist that ARE_A WineGrape?"?

No, since this is really an taxonomy this example (as you would say: ) "misuse" IS_A as a NARROWER/BROADER relation and INSTANCE_OF as an ABOUT relation. 

>In order to be able to ask "is_about" you would need an

>ObjectProperty beforehand to relate two things plus a class

>definition for "Article"...

>...To be honest, this is such an obvious misuse of not only

>subClassOf but of the whole OWL language that I don't think

>that modelers will really go that way.

As you say one can express the required relations in the standard OWL way using proper ontogical constructs based on descriptive logic. I think we all agree that one should do that if the intend is to produce ontologies that allow software to interoperate (for the semantic web). 

However for the taxonomist, I think that what you suggest would be more complex and more expensive (*).... 

In particular, my key problem with your ontological approach is that it appears not to provide out-of-the-box reasoner support for my classification question: "which (articles) exist that are about WineGrape?" (If I am wrong about the reasoner support for your apprach, please let me know - I would be very interested!!) 

Note that with the taxonomy approach out-of-the-box reasoner support does exist, since the reasoner will propagate instances (informantion object) UP(!!) the class (topic/subject) inheritance chain. This makes is possible to give correct answers to all needed taxonomy and classification questions.

(*) This is important as many taxonomists do not have a background in mathematics or computer science AND very few companies with just the need for a taxonomy for their information will be willing to spend more than necessary in order to satisfy academic concerns.

>And it will be hard

>to use a "commonly available" OWL editor as well: imagine

>you want to add separate class for books and websites (e.g.

>in order to automatically generate a link to amazon or a

>link to the website, depending on the type of the information

>object). Where do you add these classes in your taxonomy?:

 

One way would be to use facet classification, so we still have: 

- Grape

- WineGrape

- CabernetSauvignonGrape (3 instances below)

+ RestaurantThatServesCabernetSauvignon

+ BookAboutCabernetSauvignonGrape

+ WebSiteAboutCabernetSauvignonGrape 

AND 

- Object

- Book (1 instance below)

+ BookAboutCabernetSauvignonGrape

AND

- Object

- Website (1 instance below)

+ WebSiteAboutCabernetSauvignonGrape

 

That is the traditional way to do it. It should cause no problems for any OWL tools including reasoners (as long as it restricts itself to taxonomies only).

>You might say: "Bad example. 

Yes :-)

>I think we won't get a problem either because if you only

>have one type of instance, then your taxonomy is valid.

>You just labeled your classes in a way that leads to

>confusion. Rewritten taxonomy:

>

>- InfoObject

>InfoObjectAboutGrape

>InfoObjectAboutWineGrape

>InfoObjectAboutCabernetSauvignonGrape

>...

>

>All the way nice "isSublassOf" relations...

The labels are written as they typical would be in a taxonomy.

Your point about renaming is interesting, but consider what will happen when a reasoner runs on a mix of my OWL taxonomy and a OWL wine ontology.



Greetings,

M. Christensen


------------
Yahoo! Mail - Gratis: 6 MB lagerplads, spamfilter og virusscan