Display of entailments

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

Display of entailments

Bijan Parsia-3
In the current build, I'm seeing far fewer entailments in the entity  
pane than I want and am used to :)

In particular, I really want inferred disjointness, type on  
individuals, and relations between named individuals. I'd like these  
to show up in the entity pane (right now type inferences don't!) and  
in the list of inferred axioms.

Ideally, I'd like to be able to save DL Queries and even have them  
show up in the other panels.

So, for example, I might have the DL Query:

        not P some C (superclasses)

I don't want to name this, I just want to see it. Now suppose I have  
five or six such dl queries. First I would like not to have to  
retyped them, but more importantly, I want these computed every time  
and to appear inline and in the inferred list.



Cheers,
Bijan.

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

Re: Display of entailments

Nick Drummond
Hi Bijan,

Yes, some of the sections don't show inferred information - particularly in the property view. I'll add this to the list.
Inferred types of individuals do show up in the individual description, class description and inferred axioms views - if you have an example where they don't that would be useful.

Showing unnamed expressions seems reasonable enough - they can be shown in the inferred superclasses and types (and any other section that takes classes).
A history of the expressions entered in the DLQuery tab could be retained (and saved with the ontology?).
What do you mean by "appear inline"?
Added to the list.

Nick

On Thu, Jul 3, 2008 at 5:14 PM, Bijan Parsia <[hidden email]> wrote:
In the current build, I'm seeing far fewer entailments in the entity
pane than I want and am used to :)

In particular, I really want inferred disjointness, type on
individuals, and relations between named individuals. I'd like these
to show up in the entity pane (right now type inferences don't!) and
in the list of inferred axioms.

Ideally, I'd like to be able to save DL Queries and even have them
show up in the other panels.

So, for example, I might have the DL Query:

       not P some C (superclasses)

I don't want to name this, I just want to see it. Now suppose I have
five or six such dl queries. First I would like not to have to
retyped them, but more importantly, I want these computed every time
and to appear inline and in the inferred list.



Cheers,
Bijan.

_______________________________________________
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: Display of entailments

Bijan Parsia-3
On 17 Jul 2008, at 12:21, Nick Drummond wrote:

> Hi Bijan,
>
> Yes, some of the sections don't show inferred information -  
> particularly in the property view. I'll add this to the list.
> Inferred types of individuals do show up in the individual  
> description, class description and inferred axioms views - if you  
> have an example where they don't that would be useful.

Will send.

> Showing unnamed expressions seems reasonable enough - they can be  
> shown in the inferred superclasses and types (and any other section  
> that takes classes).
> A history of the expressions entered in the DLQuery tab could be  
> retained (and saved with the ontology?).

Ooo, that would be extra nice. Even over the course of a session  
would be useful.

I guess an easy way to save them is to add them as special  
definitional axioms. Or just in a big blob somewhere :)

> What do you mean by "appear inline"?
[snip]

In the entity panes like other "inferred anonymous classes".

Here's my work flow. I write some axioms. They aren't enough to force  
a subsumption or an unsatisfiability (which is ok), but I'm not 100%  
I got them right. So I go spelunking in the DL Query pane. That's  
fine for small and few tests, but as I get to much more complex  
expressions (and many alternative ones) it all falls apart.

I could create a new ontology for the "extra tests", add my ad hoc  
tests as definitions, then import my real ontology, *then* switch  
between the two...and I can also drill a hole in my knee and fill it  
with jam (as Uli likes to say :)). The workflow I'd like to have is:  
1) write some axioms, 2) explore stuff with DL Queries, 3)  
occasionally save a query as interesting or active, and 4) have that  
expression show up in all the places other, more standard, entailment  
show up (entity panes, the inferred axiom list, etc.)

This is very similar to various unit test mutterings floating about.

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