[Pellet-users] More issues with Pellet-1,4

[Evren Sirin]

I've checked this problem and it turns out to be a bug related to 
enumerated datatypes. It is fixed in the SVN now.

Regards,
Evren

On 2/26/07 6:42 PM, Cartik Saravana Muthu wrote:
> Hi Evren,
>
> I'm using an ontology with Pellet to guide instance based 
> classification. I'm sending you both the source code and the ontology 
> in question. The API's I'm using are Jena 2.5 with Pellet 1.4 . The 
> ontology has been written in OWL 1.0. I have created a taxonomy of 
> concepts, with a concept being distinguished from its siblings by "has 
> Value" restrictions on a datatype property. For example, Granulocytes 
> are distinguished from Agranulocytes by the "has granules" datatype 
> property, which takes the value "true" for Granulocytes and "false" 
> for Agranulocytes.  I had written to you a few months ago about the 
> inability of Pellet 1.3 to work with "has value" restrictions on 
> datatype properties, specifically the one mentioned above.
>
> Now, with Pellet 1.4, this seems to be working fine, except I've run 
> into a variation of the same problem. In the ontology, Eosinophils are 
> defined as Granulocytes for whom the "granules stain" datatype 
> property takes the value reddish-orange. Similarly, Basophils have 
> granules that stain blue-black and Neutrophils have granules that 
> stain lilac. Now when I load up these definitions along with an 
> instance into a Pellet 1.4 reasoner, the reasoner infers the instance 
> to be an Eosinophil irrespective of what value is asserted to the 
> granules stain property. Am I doing something wrong here? Or is this a 
> bug?
>
> Thanks in advance!!
>
> Regards,
>
> -------------------------------------------------------------------------------------------------------------------
> Cartik R. Kothari, Ph.D.
> Post Doctoral Research Fellow
> James Hogg iCAPTURE Institute of Cardiovascular and Pulmonary Research
> St. Paul's Hospital, University of British Columbia
> Vancouver, BC V6Z 1Y6