Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[protege-discussion] Question on Self-Joins to an Entity

Samson Tu swt at stanford.edu
Wed Jan 17 22:55:06 PST 2007


A join is not a meaningful operation on a class.

Having an instance x of Foo class with a(x)=a1 and b(x)=b1 is equivalent 
to the logical sentence Foo(x) and a(x, a1) and b(x, b1), where Foo is a 
unary predicate and a, b are binary predicates. You don't do joins with 
logical sentences.

Samson


Kent Andrews wrote:
> Hi, 
> 
>  
> 
> Hoping some one can share some insight into how they would or have modeled a situation like this:
> 
>  
> 
> If you have a self join on an entity which only has extra attributes applicable when the
> 
> join is in effect how do you model that?
> 
>  
> 
> For example:
> 
>  
> 
> Class Foo 
> 
> -> has attributes a & b
> 
>  
> 
> Class Foo Related to Class Foo
> 
> -> has attributes a, b, c & d
> 
>  
> 
> If I'm not making it clear I apologize. I am trying to understand how a Concept or Class within Protégé can have 2 slots
> 
>  
> 
> The problem restated is Class Foo has two slot's (slot a & slot b).
> 
> Now Class Foo can be joined to itself and once it is in that state it then has two extra attributes (slots) which it doesn't have if the self join doesn't exist.
> 
>  
> 
> Do you create a Sub-Class of Foo and give it a name like Related-Foo and upon the sub-class you stick the two extra attributes.
> 
>  
> 
> Then in Foo I have a slot which is an instance of Related-Foo with cardinality 0 or 1 ?
> 
>  
> 
> Please help any ideas much appreciated.
> 
>  
> 
> Thanks in advance.
> 
>  
> 
> Kent Andrews
> 
>  
> 
> _______________________________________________
> protege-discussion mailing list
> protege-discussion at lists.stanford.edu
> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
> 
> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03 
> 


-- 
Samson Tu                    email: swt at stanford.edu
Senior Research Scientist    web: www.stanford.edu/~swt/
Stanford Medical Informatics phone: 1-650-725-3391
Stanford University          fax: 1-650-725-7944




More information about the protege-discussion mailing list