Search Mailing List Archives
[protege-owl] Please give me advices for differences of inference between 4.0 and 3.4.1
Cho, Hyun Sang
hscho at nict.go.jp
Sun Oct 12 03:11:17 PDT 2008
I just started to study ontology and tried to use Protege for handling OWL ontology now.
I built up a simple ontology to infer simple individual type by using value assigned data property.
In fact it was same with an example on pp. 82, ch#5 in “A Practical Guide To Building OWL Ontologies Using Protege 4 and CO-CODE Tools Edition 1.1”.
The ontology was built by Protege 4.0.
I defined EnergyZoneState as an individual of class CurrentEnergyZone and assigned data property hasEnergyVaule for assertion. Just as the example in the text, if hasEnergyVaule exceeds 12, EnergyZoneState will have another type class of RedEnergyZone.
I assigned value to hasEnergyVaule and got expected results by running Pellet in menubar of Protege 4.0.
At same time, other individuals that only have hasEnergyVaule data property also inferred to have their suitable state.
So, I think, ontology is normal and operation of inference engine is also normal.
Next, I tried to do this operation by using Protege OWL API with Java (In this case, I used Protege 3.4 beta build 506 because when I opened org.semanticweb.owl.owlapi.jar in plugin folder of 4.0 I found the packages were totally different from example of inference API.)
Although I got some problems, I could run the inference operation.
However, for EnergyZoneState individual, it only returned CurrentEnergyZone as a type class.
I also tried to find inferred individuals of GreenEnergyZone but there is no individual.
So, I concluded that I got no inferred result.
I tried to check inference by using 3.4 and also couldn’t get inferred type.
Is the result limitation of 3.4? Is there no way to get the same result of 4.0?
Please give me advice to find a way to get same result with the Protege 4.0.
Cho, Hyun Sang
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the protege-owl