[protege-owl] about InterestingPizza queston

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

[protege-owl] about InterestingPizza queston

Weather-2
Hi

I'm learning how to use Protégé OWL and do the Pizza  lab
I want to find the pizzas with just three pizzatoppongs
so i change the InterestingPizza's restriction from  hasTopping > 3 to  hasTopping = 3
But it seems not reasoning by racer and other reasoning tools(like pellet-1.3)
Why ??

Protege 3.1.1 Build 216

Weather

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] about InterestingPizza queston

Thomas Russ

On Nov 29, 2006, at 10:07 PM, weather ho wrote:

> Hi
>
> I'm learning how to use Protégé OWL and do the Pizza  lab
> I want to find the pizzas with just three pizzatoppongs
> so i change the InterestingPizza's restriction from  hasTopping > 3  
> to  hasTopping = 3
> But it seems not reasoning by racer and other reasoning tools(like  
> pellet-1.3)
> Why ??

Most likely is because of open world reasoning.

Just because a given individual has only 3 toppings asserted,
it doesn't mean there aren't additional toppings that the
system hasn't been told about.  Unless there is a restriction
that indicates that the set of toppings is the full set, you
won't be able to conclude anything about maximum cardinality
just by counting fillers.


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 
Reply | Threaded
Open this post in threaded view
|

[protege-owl] Another Pizza question

JMiller

The "about InterestingPizza question" seems to be remotely similar to an issue that we have been grappling with.

Assume it is given that we have Pizza8 with 8 toppings, Pizza5 with 5 toppings, and Pizza3 with 3 toppings.  Some of the toppings are easily identified, while others (maybe some spices) are difficult to detect.  The 3 toppings in Pizza3 are included in Pizza8's toppings.  Pizza5 shares only 2 toppings with Pizza3 and Pizza8.

We are given a pizza, and we can identify 3 toppings, all of which are toppings contained in Pizza3 (and therefore, also Pizza8), and only 2 of them are in Pizza5.

 I think we would want to get the result:  This is definitely not a Pizza5, but could be a Pizza3 or Pizza8 (but it is an exact match of Pizza3, only a partial match of Pizza8).   (It seems that the Open World would say that this could be a Pizza3 or Pizza8 with equal likelihood, but I'm still trying to fully grasp OWA).

What inferences can be made here, if any, about what kind of pizza we have?  And would these be determined by (some combination of) restrictions, rules, or other?  How does Open World relate to this, if at all?  What could one normally expect to receive from a reasoner in a case like this?

Jim Miller


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] about InterestingPizza queston

Timothy Redmond
In reply to this post by Weather-2

A simple test suggests that this works.  I use pellet (it is easily  
obtained).  I made a small ontology where I defined A to be "f  
exactly 3" and I defined B to be "f min 2".  Running pellet easily  
determined that A was a subclass of B.

What is making you think that inference is not working?  Or more  
precisely, what inferences are not being done?

-Timothy



On Nov 29, 2006, at 10:07 PM, weather ho wrote:

> Hi
>
> I'm learning how to use Protégé OWL and do the Pizza  lab
> I want to find the pizzas with just three pizzatoppongs
> so i change the InterestingPizza's restriction from  hasTopping > 3  
> to  hasTopping = 3
> But it seems not reasoning by racer and other reasoning tools(like  
> pellet-1.3)
> Why ??
>
> Protege 3.1.1 Build 216
>
> Weather
> _______________________________________________
> protege-owl mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-owl
>
> Instructions for unsubscribing: http://protege.stanford.edu/doc/ 
> faq.html#01a.03

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] about InterestingPizza queston

Weather-2
In reply to this post by Thomas Russ
Hi

how to do the full set ?
Like Mushroom Pizza  ==>. ∀ hasTopping (MozzarellaTopping ⊔ MushroomTopping ⊔ TomatoTopping) or the other way ??

And i must explain why i do the change about  InterestingPizza queston hasTopping  =  3
Because i am try to do the mobile phone's owl and i got a question as following

Mobile phone has DualBand frequence or TriBand property frequence so i want to taxonomy this kind of class .

thanks


2006/12/1, Thomas Russ <[hidden email]>:

Most likely is because of open world reasoning.

Just because a given individual has only 3 toppings asserted,
it doesn't mean there aren't additional toppings that the
system hasn't been told about.  Unless there is a restriction
that indicates that the set of toppings is the full set, you
won't be able to conclude anything about maximum cardinality
just by counting fillers.


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 
Reply | Threaded
Open this post in threaded view
|

Re: [protege-owl] Another Pizza question

JMiller
In reply to this post by JMiller

I am re-submitting this question, on the chance it was overlooked.

Jim Miller




James A Miller <[hidden email]>
Sent by: [hidden email]

11/30/2006 01:08 PM

Please respond to
User support for the Protege-OWL editor        <[hidden email]>

To
User support for the Protege-OWL editor <[hidden email]>
cc
Subject
[protege-owl] Another Pizza question






The "about InterestingPizza question" seems to be remotely similar to an issue that we have been grappling with.


Assume it is given that we have Pizza8 with 8 toppings, Pizza5 with 5 toppings, and Pizza3 with 3 toppings.  Some of the toppings are easily identified, while others (maybe some spices) are difficult to detect.  The 3 toppings in Pizza3 are included in Pizza8's toppings.  Pizza5 shares only 2 toppings with Pizza3 and Pizza8.


We are given a pizza, and we can identify 3 toppings, all of which are toppings contained in Pizza3 (and therefore, also Pizza8), and only 2 of them are in Pizza5.


I think we would want to get the result:  This is definitely not a Pizza5, but could be a Pizza3 or Pizza8 (but it is an exact match of Pizza3, only a partial match of Pizza8).   (It seems that the Open World would say that this could be a Pizza3 or Pizza8 with equal likelihood, but I'm still trying to fully grasp OWA).


What inferences can be made here, if any, about what kind of pizza we have?  And would these be determined by (some combination of) restrictions, rules, or other?  How does Open World relate to this, if at all?  What could one normally expect to receive from a reasoner in a case like this?


Jim Miller

_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03


_______________________________________________
protege-owl mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-owl

Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03