Bug in DL Query: instances

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Bug in DL Query: instances

Alan Ruttenberg-2
Create a class A.
Make B a subclass.
Create an individual b1, instance of B
Classify
Do a DL Query with the query being A
check off instances.
b1 is not displayed, though it should be. (if b1 type B, and B
subClassOf A, then b1 type A)
If the query is for B then b1 is correctly shown.

-Alan
_______________________________________________
p4-feedback mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/p4-feedback
Reply | Threaded
Open this post in threaded view
|

Re: Bug in DL Query: instances

regmito
This is a bug in Protégé GUI, not in the reasoner. So, if the DL query is executed programmatically (outside Protege), then is works fine. This bug has been introduced in Protégé 4.1, while Protégé 4.0.2 build 115 works fine with this query. There is another thread in this mail list about this bug, (I think) discussed a couple of weeks ago.

Cheers,
Igor

Dr Igor Toujilov
Senior Research Fellow
Centre for Health Informatics and Multiprofessional Education
University College London
Highgate Hill
London, UK
N19 5LW

Tel: +44 (0)20 7288 3960

-----Original Message-----
From: [hidden email] [mailto:[hidden email]] On Behalf Of Alan Ruttenberg
Sent: 22 October 2010 07:36
To: Protege 4 feedback
Subject: [p4-feedback] Bug in DL Query: instances

Create a class A.
Make B a subclass.
Create an individual b1, instance of B
Classify
Do a DL Query with the query being A
check off instances.
b1 is not displayed, though it should be. (if b1 type B, and B
subClassOf A, then b1 type A)
If the query is for B then b1 is correctly shown.

-Alan
_______________________________________________
p4-feedback mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/p4-feedback
_______________________________________________
p4-feedback mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/p4-feedback
Reply | Threaded
Open this post in threaded view
|

Re: Bug in DL Query: instances

Timothy Redmond
In reply to this post by Alan Ruttenberg-2

Get the latest Protege and the latest updates and you will see the behavior that you want.

I would point out that the behavior old version is the same as the behavior that you see for classes.

  1. Create a class A
  2. Create a subclass B of A
  3. Create a subclass C of B
  4. Start the reasoner
  5. do a dlquery of a for subclasses

C is not displayed even though following your logic it should be.

Some work still needs to be done on the dlquery tab (and some of the other displayed inferences) for people who have lots of individuals.  I suspect that most of those people are using Protege 3 though because of the forms issue.

-Timothy


On 10/21/2010 11:36 PM, Alan Ruttenberg wrote:
Create a class A.
Make B a subclass.
Create an individual b1, instance of B
Classify
Do a DL Query with the query being A
check off instances.
b1 is not displayed, though it should be. (if b1 type B, and B
subClassOf A, then b1 type A)
If the query is for B then b1 is correctly shown.

-Alan
_______________________________________________
p4-feedback mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/p4-feedback
  


_______________________________________________
p4-feedback mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/p4-feedback
Reply | Threaded
Open this post in threaded view
|

Re: Bug in DL Query: instances

Alan Ruttenberg-2
On Fri, Oct 22, 2010 at 12:45 PM, Timothy Redmond <[hidden email]> wrote:
>
> Get the latest Protege and the latest updates and you will see the behavior
> that you want.
Thanks.

>
> I would point out that the behavior old version is the same as the behavior
> that you see for classes.
>
> Create a class A
> Create a subclass B of A
> Create a subclass C of B
> Start the reasoner
> do a dlquery of a for subclasses
>
> C is not displayed even though following your logic it should be.

Yes. However it is possible to get the behavior I want using descendants.

That said, I think the labels on the checkboxes should be changed.

"Subclasses" -> "Direct subclasses" or "Immediate subclasses" or "Most
general subclasses"
"Super Classes" - "Direct superclasses" or "Immediate superclasses" or
"Most specific superclasses"

(probably the last of the three is the most informative and least jargony)

"Descendant classes" -> "Subclasses"
"Ancestor classes" -> "Superclasses"

If you think the current behavior on individuals is useful, then
include a checkbox for that. I can't think of a short name for this.
They might be called "Direct instances" but I don't really like that.
They are individuals that aren't also individuals of a more specific
class.

-Alan

>
> Some work still needs to be done on the dlquery tab (and some of the other
> displayed inferences) for people who have lots of individuals.  I suspect
> that most of those people are using Protege 3 though because of the forms
> issue.
>
> -Timothy
>
>
> On 10/21/2010 11:36 PM, Alan Ruttenberg wrote:
>
> Create a class A.
> Make B a subclass.
> Create an individual b1, instance of B
> Classify
> Do a DL Query with the query being A
> check off instances.
> b1 is not displayed, though it should be. (if b1 type B, and B
> subClassOf A, then b1 type A)
> If the query is for B then b1 is correctly shown.
>
> -Alan
> _______________________________________________
> p4-feedback mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/p4-feedback
>
>
> _______________________________________________
> p4-feedback mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/p4-feedback
>
>
_______________________________________________
p4-feedback mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/p4-feedback