[Pellet-users] Least Common Ancestor Bug?

[Jason Rickwald]

Hello,

I'm in the process of playing with the different reasoning mechanisms 
available in Pellet to figure out how we can use it for an OWL-related 
project.  I was testing the method computeLCA( ATermList ) in the 
Taxonomy class to compute the least common ancestor for a set of 
classes.  However, I can't seem to get it to work as I would expect it 
to.  For example, with an ontology like:

Thing
   |
   - Animal
   |     |
   |     - Person
   - Plant

If I were to pass in Person and Plant, I would expect to get Thing back, 
but I don't get anything.

I looked at the source code and found a possible bug -- calls to 
getSupers( ATermAppl, direct, flat ) have direct set to true.  I tried 
"fixing" it by setting direct to false and got behavior more like what I 
was expecting.  It still doesn't work quite right, though.  For example, 
if Person had subclasses Man and Woman, and they were passed into the 
"fixed" computLCA, it would return Animal instead of Person.  I'm 
looking into why this would be.

Before I go any further with this, though, I want to make sure that 
getting only direct classes _is_ a bug and not merely my own 
misunderstanding of how this method is supposed to work.  Thanks for 
your help.

-Jason
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 5209 bytes
Desc: S/MIME Cryptographic Signature
Url : http://lists.owldl.com/pipermail/pellet-users/attachments/20070116/d6ce9f36/attachment.bin