Medical report

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

Medical report

Jabardi
Hello all,
Suppose I have a medical records that has a person and disease Classes
The class person has two sub-class healthy and sick
We consider that I have 1,000 instances. Each instance is Healthy or Sick.
 When I used a SWRL rule to infer sick persons and I got 300 sick insatnces.

How can I get the rest instances healthy persons

Regards
Mohammed



--
Sent from: http://protege-project.136.n4.nabble.com/Protege-User-f4659818.html
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user
Reply | Threaded
Open this post in threaded view
|

Re: Medical report

Martin O'Connor-2

It appears that you may be attempting to perform closed-world reasoning here. 

See:

https://github.com/protegeproject/swrlapi/wiki/SWRLLanguageFAQ#does-swrl-adopt-the-open-world-assumption

Martin

On Jul 19, 2019, at 7:21 AM, Jabardi <[hidden email]> wrote:

Hello all,
Suppose I have a medical records that has a person and disease Classes
The class person has two sub-class healthy and sick
We consider that I have 1,000 instances. Each instance is Healthy or Sick.
When I used a SWRL rule to infer sick persons and I got 300 sick insatnces.

How can I get the rest instances healthy persons

Regards
Mohammed



--
Sent from: http://protege-project.136.n4.nabble.com/Protege-User-f4659818.html
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user


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

Re: Medical report

Jabardi

Thanks Martin.

I meant how express (belong to, not belong to and not in) in SWRL

 

 

 

From: [hidden email]
Sent: Friday, July 19, 2019 6:06 PM
To: [hidden email]
Subject: Re: [protege-user] Medical report

 

 

It appears that you may be attempting to perform closed-world reasoning here. 

 

See:

 

https://github.com/protegeproject/swrlapi/wiki/SWRLLanguageFAQ#does-swrl-adopt-the-open-world-assumption

 

Martin



On Jul 19, 2019, at 7:21 AM, Jabardi <[hidden email]> wrote:

 

Hello all,
Suppose I have a medical records that has a person and disease Classes
The class person has two sub-class healthy and sick
We consider that I have 1,000 instances. Each instance is Healthy or Sick.
When I used a SWRL rule to infer sick persons and I got 300 sick insatnces.

How can I get the rest instances healthy persons

Regards
Mohammed



--
Sent from: http://protege-project.136.n4.nabble.com/Protege-User-f4659818.html
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user

 

 


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

Re: Medical report

Lorenz Buehmann
That's why he shared the link with you. You can't, OWL and SWRL use Open World Assumption.

Saying something like

not sick SubClassOf healthy

doesn't work. Please read about Open World assumption to understand why absence of information can't be considered as negation during inference.



Thanks Martin.

I meant how express (belong to, not belong to and not in) in SWRL

 

 

 

From: [hidden email]
Sent: Friday, July 19, 2019 6:06 PM
To: [hidden email]
Subject: Re: [protege-user] Medical report

 

 

It appears that you may be attempting to perform closed-world reasoning here. 

 

See:

 

https://github.com/protegeproject/swrlapi/wiki/SWRLLanguageFAQ#does-swrl-adopt-the-open-world-assumption

 

Martin



On Jul 19, 2019, at 7:21 AM, Jabardi <[hidden email]> wrote:

 

Hello all,
Suppose I have a medical records that has a person and disease Classes
The class person has two sub-class healthy and sick
We consider that I have 1,000 instances. Each instance is Healthy or Sick.
When I used a SWRL rule to infer sick persons and I got 300 sick insatnces.

How can I get the rest instances healthy persons

Regards
Mohammed



--
Sent from: http://protege-project.136.n4.nabble.com/Protege-User-f4659818.html
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user

 

 


_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user
-- 
Lorenz Bühmann
AKSW group, University of Leipzig
Group: http://aksw.org - semantic web research center

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

Re: Medical report

Igor Toujilov-2

Hi, Mohammed.

This means: you should explicitly assert facts, which make healthy persons healthy. This is similar to how you explicitly assert facts about person’s disease. It is because OWL treats positive facts and negative facts the same way.

 

Cheers,

Igor

 

 
Sent: Sunday, July 21, 2019 at 8:24 AM
From: "Lorenz B." <[hidden email]>
To: [hidden email]
Subject: Re: [protege-user] Medical report
That's why he shared the link with you. You can't, OWL and SWRL use Open World Assumption.
 
Saying something like
 
not sick SubClassOf healthy
 
doesn't work. Please read about Open World assumption to understand why absence of information can't be considered as negation during inference.
 
 
 

Thanks Martin.

I meant how express (belong to, not belong to and not in) in SWRL

 

 

 

From: [hidden email]
Sent: Friday, July 19, 2019 6:06 PM
To: [hidden email]
Subject: Re: [protege-user] Medical report

 

 

It appears that you may be attempting to perform closed-world reasoning here. 

 

See:

 

https://github.com/protegeproject/swrlapi/wiki/SWRLLanguageFAQ#does-swrl-adopt-the-open-world-assumption

 

Martin


 

On Jul 19, 2019, at 7:21 AM, Jabardi <[hidden email]> wrote:

 

Hello all,
Suppose I have a medical records that has a person and disease Classes
The class person has two sub-class healthy and sick
We consider that I have 1,000 instances. Each instance is Healthy or Sick.
When I used a SWRL rule to infer sick persons and I got 300 sick insatnces.

How can I get the rest instances healthy persons

Regards
Mohammed



--
Sent from: http://protege-project.136.n4.nabble.com/Protege-User-f4659818.html
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user

 

 

 
 
_______________________________________________
protege-user mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-user
--
Lorenz Bühmann
AKSW group, University of Leipzig
Group: http://aksw.org - semantic web research center
_______________________________________________ protege-user mailing list [hidden email] https://mailman.stanford.edu/mailman/listinfo/protege-user

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