Bayes

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

Bayes

Mirzaoff, Alexander

Is it possible to organize a Protégé model into a Bayes network? Where could I search such an application in the Protégé documentation online?

Thanks for any help,

Alex

 


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

Re: Bayes

Michael DeBellis-2
Alex, it's an interesting question but it's an open research question not something that is implemented in Protege. In standard OWL (the language Protege uses to model domains) there is no way to associate a probability with an axiom or concept. Things are either just true or false (or not known).   If you Google "OWL Bayes Network" you will get lots of results. Many people seem to be working on this. Here are 3 that I thought looked interesting but there are many more:




Michael



On Thu, Nov 9, 2017 at 12:00 PM, Mirzaoff, Alexander <[hidden email]> wrote:

Is it possible to organize a Protégé model into a Bayes network? Where could I search such an application in the Protégé documentation online?

Thanks for any help,

Alex

 


_______________________________________________
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: [Marketing] Re: Bayes

Mirzaoff, Alexander

Thanks much Michael.

Alex

 

From: protege-user [mailto:[hidden email]] On Behalf Of Michael DeBellis
Sent: Thursday, November 09, 2017 5:16 PM
To: User support for WebProtege and Protege Desktop <[hidden email]>
Subject: [Marketing] Re: [protege-user] Bayes

 

Alex, it's an interesting question but it's an open research question not something that is implemented in Protege. In standard OWL (the language Protege uses to model domains) there is no way to associate a probability with an axiom or concept. Things are either just true or false (or not known).   If you Google "OWL Bayes Network" you will get lots of results. Many people seem to be working on this. Here are 3 that I thought looked interesting but there are many more:

 

 

 

 

Michael

 

 

 

On Thu, Nov 9, 2017 at 12:00 PM, Mirzaoff, Alexander <[hidden email]> wrote:

Is it possible to organize a Protégé model into a Bayes network? Where could I search such an application in the Protégé documentation online?

Thanks for any help,

Alex

 


_______________________________________________
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: [Marketing] Re: Bayes

Jim McCusker
We implemented probabilities on general RDF data by encapsulating it in nanopublications. [1] We didn't do this with OWL axioms, but you can see a potential encoding there. How that is translated into a bayes net is up to you. 

[1] McCusker JP, Dumontier M, Yan R, He S, Dordick JS, McGuinness DL. (2017) Finding melanoma drugs through a probabilistic knowledge graph. PeerJ Computer Science 3:e106 https://doi.org/10.7717/peerj-cs.106

On Thu, Nov 9, 2017 at 5:20 PM Mirzaoff, Alexander <[hidden email]> wrote:

Thanks much Michael.

Alex

 

From: protege-user [mailto:[hidden email]] On Behalf Of Michael DeBellis
Sent: Thursday, November 09, 2017 5:16 PM
To: User support for WebProtege and Protege Desktop <[hidden email]>
Subject: [Marketing] Re: [protege-user] Bayes

 

Alex, it's an interesting question but it's an open research question not something that is implemented in Protege. In standard OWL (the language Protege uses to model domains) there is no way to associate a probability with an axiom or concept. Things are either just true or false (or not known).   If you Google "OWL Bayes Network" you will get lots of results. Many people seem to be working on this. Here are 3 that I thought looked interesting but there are many more:

 

 

 

 

Michael

 

 

 

On Thu, Nov 9, 2017 at 12:00 PM, Mirzaoff, Alexander <[hidden email]> wrote:

Is it possible to organize a Protégé model into a Bayes network? Where could I search such an application in the Protégé documentation online?

Thanks for any help,

Alex

 


_______________________________________________
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
--
James P. McCusker III, Ph.D.

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

Re: [Marketing] Re: [Marketing] Re: Bayes

Mirzaoff, Alexander

Thankyou Jim. This sounds very interesting, particularly the concept of nanopublications, and the ingest into SPARQL. I’ve got the paper and am in the study mode.

This is some outstanding work.

 

From: protege-user [mailto:[hidden email]] On Behalf Of Jim McCusker
Sent: Thursday, November 09, 2017 5:24 PM
To: User support for WebProtege and Protege Desktop <[hidden email]>
Subject: [Marketing] Re: [protege-user] [Marketing] Re: Bayes

 

We implemented probabilities on general RDF data by encapsulating it in nanopublications. [1] We didn't do this with OWL axioms, but you can see a potential encoding there. How that is translated into a bayes net is up to you. 

 

[1] McCusker JP, Dumontier M, Yan R, He S, Dordick JS, McGuinness DL. (2017) Finding melanoma drugs through a probabilistic knowledge graph. PeerJ Computer Science 3:e106 https://doi.org/10.7717/peerj-cs.106

 

On Thu, Nov 9, 2017 at 5:20 PM Mirzaoff, Alexander <[hidden email]> wrote:

Thanks much Michael.

Alex

 

From: protege-user [mailto:[hidden email]] On Behalf Of Michael DeBellis
Sent: Thursday, November 09, 2017 5:16 PM
To: User support for WebProtege and Protege Desktop <[hidden email]>
Subject: [Marketing] Re: [protege-user] Bayes

 

Alex, it's an interesting question but it's an open research question not something that is implemented in Protege. In standard OWL (the language Protege uses to model domains) there is no way to associate a probability with an axiom or concept. Things are either just true or false (or not known).   If you Google "OWL Bayes Network" you will get lots of results. Many people seem to be working on this. Here are 3 that I thought looked interesting but there are many more:

 

 

 

 

Michael

 

 

 

On Thu, Nov 9, 2017 at 12:00 PM, Mirzaoff, Alexander <[hidden email]> wrote:

Is it possible to organize a Protégé model into a Bayes network? Where could I search such an application in the Protégé documentation online?

Thanks for any help,

Alex

 


_______________________________________________
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

--

James P. McCusker III, Ph.D.


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