Differences in displayed inferred reasoner output

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

Differences in displayed inferred reasoner output

Gibson, A.P.
Differences in displayed inferred reasoner output

Hi Nick et al,

Classifying under Pellet gives nice yellowish dashed boxes of inferred output - this is especially useful for quickly verifying inferred domains, ranges and inverses of properties for example, and also of inferred properties on individuals.

Classifying with Fact++ does not show (m)any of these inferences. I can find out if they are there currently by checking them using the DL-Query tab, but I have to pick examples of inferences I would expect if the inference were true and match the result to it. I have found more in depth differences between the inferences made by the two reasoners (where Pellet seems to be misbehaving) and I would like to remove the ambiguity between by what inferences just do and do not get displayed in Protege by the two reasoners, and what inferences are genuinely different between the two reasoners.

I apologise if this is something that is on your list Nick, but its quite a major one as currently it is made difficult for anyone to work out if a) Protege is not working as one would expect b) The resoners are not working as one expect or are at least are producing genuinely different inferences or c) if there is something wrong with the modelling.


Dr Andrew Gibson
Universiteit van Amsterdam

p4-feedback mailing list
[hidden email]