How to get customized results when solving the Einstein puzzle with OWL in Protégé?

I would like to better understand ontologies and reasoning. There is an interesting conundrum of Einstein's riddle online that can be solved with ontologies and reasoning. I downloaded the OWL ontology from this site and imported it into Protege 4.0.2 (not working with 4.1). I can start reasoning Reasoner -> FaCT ++, Reasoner -> Classify ... but I don't know how to visualize the individual results. How can i do this?


source to share

2 answers

There are two ways to visualize the results. First, when you select the Classes tab, you will see that there are two views available to you: the class hierarchy view, which is the declared taxonomy, and the class hierarchy view (output), which, as the name suggests, is inferred. This gives you an idea of ​​the class. As Kaarel suggests, you can visualize people reasoning in the Individuals tab.

Protege also lets you add multiple other faces to the user interface: create a new blank tab and then choose a view from the Individuals submenu in the Views menu. Finally, you can use a number of external graphical rendering tools: if you assert the inferred hierarchy and save it to a file (in RDF-XML), you can use tools like Welkin , IsaViz , etc. to get graphical representations.



When you are finished, go to Individuals-tab

and find the axioms with a yellow background. These axioms are enticing. If the components of the puzzle (i.e., Men, Pets, Drinks, etc.) were modeled as OWL humans, then you will see some new object property statements that the physician thinks are found between these people.

There are also other options for viewing attachments:

  • View -> Ontology views -> Classification Results

    will show a list of nested axioms. It may not display all nested axioms, although for example I tried it with Protege 4.1 and didn't see any validated object properties.
  • In DL Query tab

    you can enter a class expression and include all its subclasses and individuals (including nested ones). This may be the most natural way to study drives.

See also:



All Articles