Enterprise Data Modeling with OWL and Protege?

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

Enterprise Data Modeling with OWL and Protege?

Michael DeBellis-2
I have a potential client, a large social media company, that is building an enterprise wide data model and may be interested in using OWL and Protege as part of the project. If anyone knows of any good papers or other sources on the topic please let me know. In addition to modeling the social media domain security and integrity constraints on data are two of the most significant issues for them. I plan to look at SHACL regarding the integrity constraints, I've heard it mentioned often but have never used it. If anyone has feedback on those two topics as well please let me know. thanks,

Michael

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

Re: Enterprise Data Modeling with OWL and Protege?

Luis Enrique Ramos García
Hi Michael, 

I have worked for some time in enterprise modeling, but for manufacturing, however it is enterprise modeling, and I have work with protege, and OWL mostly.

This is one of my papers:


and I also held three editions of a workshop about enterprise modeling, here is the reference:


I am currently working in the modeling enterprises data,  and reasoning.

Best regards


Luis Ramos

El vie., 10 ene. 2020 a las 4:00, Michael DeBellis (<[hidden email]>) escribió:
I have a potential client, a large social media company, that is building an enterprise wide data model and may be interested in using OWL and Protege as part of the project. If anyone knows of any good papers or other sources on the topic please let me know. In addition to modeling the social media domain security and integrity constraints on data are two of the most significant issues for them. I plan to look at SHACL regarding the integrity constraints, I've heard it mentioned often but have never used it. If anyone has feedback on those two topics as well please let me know. thanks,

Michael
_______________________________________________
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: Enterprise Data Modeling with OWL and Protege?

Dave McComb-2
In reply to this post by Michael DeBellis-2

One of our consultants (Michael Uschold) I believe may have coined the term “enterprise ontology” in this 1996 paper

 

http://www.aiai.ed.ac.uk/publications/documents/1996/96-enterprise-ontology.pdf

 

We’ve evolved quite a bit since then, but this was an excellent start.  This is our core practice now.

 

Sent from Mail for Windows 10

 

From: [hidden email]
Sent: Thursday, January 9, 2020 8:00 PM
To: [hidden email]
Subject: [protege-user] Enterprise Data Modeling with OWL and Protege?

 

I have a potential client, a large social media company, that is building an enterprise wide data model and may be interested in using OWL and Protege as part of the project. If anyone knows of any good papers or other sources on the topic please let me know. In addition to modeling the social media domain security and integrity constraints on data are two of the most significant issues for them. I plan to look at SHACL regarding the integrity constraints, I've heard it mentioned often but have never used it. If anyone has feedback on those two topics as well please let me know. thanks,

 

Michael

 


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

Re: Enterprise Data Modeling with OWL and Protege?

Michael DeBellis-2
Dave and Luis, thanks very much for that info, I appreciate it. 

Michael

On Fri, Jan 10, 2020 at 7:21 AM Dave McComb <[hidden email]> wrote:

One of our consultants (Michael Uschold) I believe may have coined the term “enterprise ontology” in this 1996 paper

 

http://www.aiai.ed.ac.uk/publications/documents/1996/96-enterprise-ontology.pdf

 

We’ve evolved quite a bit since then, but this was an excellent start.  This is our core practice now.

 

Sent from Mail for Windows 10

 

From: [hidden email]
Sent: Thursday, January 9, 2020 8:00 PM
To: [hidden email]
Subject: [protege-user] Enterprise Data Modeling with OWL and Protege?

 

I have a potential client, a large social media company, that is building an enterprise wide data model and may be interested in using OWL and Protege as part of the project. If anyone knows of any good papers or other sources on the topic please let me know. In addition to modeling the social media domain security and integrity constraints on data are two of the most significant issues for them. I plan to look at SHACL regarding the integrity constraints, I've heard it mentioned often but have never used it. If anyone has feedback on those two topics as well please let me know. thanks,

 

Michael

 

_______________________________________________
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: Enterprise Data Modeling with OWL and Protege?

Igor Toujilov-2
In reply to this post by Michael DeBellis-2

For the last decade many people were convinced that OWL cannot handle integrity constraints properly. This is not true if you know how to do it properly.

A widespread recommendation for implementing integrity constraints is: use SHACL. I do not recommend so. SHACL is a low level language because it is RDF-based.

Please unleash the power of description logics and use OWL and SWRL for integrity constraints. The key principle is: do not allow an integrity constraint violation to drag your ontology into inconsistency. Instead just introduce a subclass of constraint violations and specify logical conditions for this subclass, so a reasoner can classify the inconsistencies.

 

Regards,

Igor

Sent: Friday, January 10, 2020 at 3:00 AM
From: "Michael DeBellis" <[hidden email]>
To: "User support for WebProtege and Protege Desktop" <[hidden email]>
Subject: [protege-user] Enterprise Data Modeling with OWL and Protege?
I have a potential client, a large social media company, that is building an enterprise wide data model and may be interested in using OWL and Protege as part of the project. If anyone knows of any good papers or other sources on the topic please let me know. In addition to modeling the social media domain security and integrity constraints on data are two of the most significant issues for them. I plan to look at SHACL regarding the integrity constraints, I've heard it mentioned often but have never used it. If anyone has feedback on those two topics as well please let me know. thanks,
 
Michael
_______________________________________________ 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: Enterprise Data Modeling with OWL and Protege?

Michael DeBellis-2
Igor, that's a very interesting comment. That was my initial reaction to SHACL as well, that I prefer the power of OWL rather than having to use RDF.  I have two questions: 

1) I think I understand what you mean when you say "introduce a subclass of constraint violations" I think you gave an example of this in reply to a recent question from a new user. But can you say any more on this, or point me to any papers or examples?

2) What about the Open World Assumption and Negation as Failure? A lot of integrity constraints involve things like saying "A Person must have a contact email" which I think can't be represented in OWL or SWRL due to OWA. Does your approach of creating classes for constraint violations provide a way around this? If so could you give an example?

Thanks for the feedback,
Michael

On Fri, Jan 10, 2020 at 5:16 PM Igor Toujilov <[hidden email]> wrote:

For the last decade many people were convinced that OWL cannot handle integrity constraints properly. This is not true if you know how to do it properly.

A widespread recommendation for implementing integrity constraints is: use SHACL. I do not recommend so. SHACL is a low level language because it is RDF-based.

Please unleash the power of description logics and use OWL and SWRL for integrity constraints. The key principle is: do not allow an integrity constraint violation to drag your ontology into inconsistency. Instead just introduce a subclass of constraint violations and specify logical conditions for this subclass, so a reasoner can classify the inconsistencies.

 

Regards,

Igor

Sent: Friday, January 10, 2020 at 3:00 AM
From: "Michael DeBellis" <[hidden email]>
To: "User support for WebProtege and Protege Desktop" <[hidden email]>
Subject: [protege-user] Enterprise Data Modeling with OWL and Protege?
I have a potential client, a large social media company, that is building an enterprise wide data model and may be interested in using OWL and Protege as part of the project. If anyone knows of any good papers or other sources on the topic please let me know. In addition to modeling the social media domain security and integrity constraints on data are two of the most significant issues for them. I plan to look at SHACL regarding the integrity constraints, I've heard it mentioned often but have never used it. If anyone has feedback on those two topics as well please let me know. thanks,
 
Michael
_______________________________________________ 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

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

Re: Enterprise Data Modeling with OWL and Protege?

Igor Toujilov-2

Hi Michael,

Here are the answers.

1. Recently I helped David in this list to formalise his model integrity, see [1]. But we did not have a concrete example in an OWL file. Now I decided to create an OWL file (attached role_integrity_constraints.owl) for other people as well. First I suggested to specify invalid_role_in_operation class for integrity violations. Now I realise it is simpler to define valid_role_in_operation instead for this use case. So, there are two roles in operation: R1 and R2. The SWLR rule classifies R1 as valid_role_in_operation, and does not classify R2 under this class. This is enough for verifying the constraint.

 

2. The second attached example, Email_presence_requirement.owl, shows how to specify in pure OWL the constraint “a person must have a contact email". There are two persons, p1 and p2, but only p1 is classifies as person_with_validated_information.

 

Cheers,

Igor

 

[1] http://protege-project.136.n4.nabble.com/Mechanism-to-ensure-model-integrity-td4673032.html

 

 
 
Sent: Sunday, January 12, 2020 at 6:11 PM
From: "Michael DeBellis" <[hidden email]>
To: "User support for WebProtege and Protege Desktop" <[hidden email]>
Subject: Re: [protege-user] Enterprise Data Modeling with OWL and Protege?
Igor, that's a very interesting comment. That was my initial reaction to SHACL as well, that I prefer the power of OWL rather than having to use RDF.  I have two questions: 
 
1) I think I understand what you mean when you say "introduce a subclass of constraint violations" I think you gave an example of this in reply to a recent question from a new user. But can you say any more on this, or point me to any papers or examples?
 
2) What about the Open World Assumption and Negation as Failure? A lot of integrity constraints involve things like saying "A Person must have a contact email" which I think can't be represented in OWL or SWRL due to OWA. Does your approach of creating classes for constraint violations provide a way around this? If so could you give an example?
 
Thanks for the feedback,
Michael
 
On Fri, Jan 10, 2020 at 5:16 PM Igor Toujilov <[hidden email]> wrote:

For the last decade many people were convinced that OWL cannot handle integrity constraints properly. This is not true if you know how to do it properly.

A widespread recommendation for implementing integrity constraints is: use SHACL. I do not recommend so. SHACL is a low level language because it is RDF-based.

Please unleash the power of description logics and use OWL and SWRL for integrity constraints. The key principle is: do not allow an integrity constraint violation to drag your ontology into inconsistency. Instead just introduce a subclass of constraint violations and specify logical conditions for this subclass, so a reasoner can classify the inconsistencies.

 

Regards,

Igor

Sent: Friday, January 10, 2020 at 3:00 AM
From: "Michael DeBellis" <[hidden email]>
To: "User support for WebProtege and Protege Desktop" <[hidden email]>
Subject: [protege-user] Enterprise Data Modeling with OWL and Protege?
I have a potential client, a large social media company, that is building an enterprise wide data model and may be interested in using OWL and Protege as part of the project. If anyone knows of any good papers or other sources on the topic please let me know. In addition to modeling the social media domain security and integrity constraints on data are two of the most significant issues for them. I plan to look at SHACL regarding the integrity constraints, I've heard it mentioned often but have never used it. If anyone has feedback on those two topics as well please let me know. thanks,
 
Michael
_______________________________________________ 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
_______________________________________________ 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

Email_presence_requirement.owl (4K) Download Attachment
role_integrity_constraints.owl (14K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Enterprise Data Modeling with OWL and Protege?

Michael DeBellis-2
Thanks Igor, I will take a look at those. I always like seeing examples. 

Cheers,
Michael

On Tue, Jan 14, 2020 at 1:43 PM Igor Toujilov <[hidden email]> wrote:

Hi Michael,

Here are the answers.

1. Recently I helped David in this list to formalise his model integrity, see [1]. But we did not have a concrete example in an OWL file. Now I decided to create an OWL file (attached role_integrity_constraints.owl) for other people as well. First I suggested to specify invalid_role_in_operation class for integrity violations. Now I realise it is simpler to define valid_role_in_operation instead for this use case. So, there are two roles in operation: R1 and R2. The SWLR rule classifies R1 as valid_role_in_operation, and does not classify R2 under this class. This is enough for verifying the constraint.

 

2. The second attached example, Email_presence_requirement.owl, shows how to specify in pure OWL the constraint “a person must have a contact email". There are two persons, p1 and p2, but only p1 is classifies as person_with_validated_information.

 

Cheers,

Igor

 

[1] http://protege-project.136.n4.nabble.com/Mechanism-to-ensure-model-integrity-td4673032.html

 

 
 
Sent: Sunday, January 12, 2020 at 6:11 PM
From: "Michael DeBellis" <[hidden email]>
To: "User support for WebProtege and Protege Desktop" <[hidden email]>
Subject: Re: [protege-user] Enterprise Data Modeling with OWL and Protege?
Igor, that's a very interesting comment. That was my initial reaction to SHACL as well, that I prefer the power of OWL rather than having to use RDF.  I have two questions: 
 
1) I think I understand what you mean when you say "introduce a subclass of constraint violations" I think you gave an example of this in reply to a recent question from a new user. But can you say any more on this, or point me to any papers or examples?
 
2) What about the Open World Assumption and Negation as Failure? A lot of integrity constraints involve things like saying "A Person must have a contact email" which I think can't be represented in OWL or SWRL due to OWA. Does your approach of creating classes for constraint violations provide a way around this? If so could you give an example?
 
Thanks for the feedback,
Michael
 
On Fri, Jan 10, 2020 at 5:16 PM Igor Toujilov <[hidden email]> wrote:

For the last decade many people were convinced that OWL cannot handle integrity constraints properly. This is not true if you know how to do it properly.

A widespread recommendation for implementing integrity constraints is: use SHACL. I do not recommend so. SHACL is a low level language because it is RDF-based.

Please unleash the power of description logics and use OWL and SWRL for integrity constraints. The key principle is: do not allow an integrity constraint violation to drag your ontology into inconsistency. Instead just introduce a subclass of constraint violations and specify logical conditions for this subclass, so a reasoner can classify the inconsistencies.

 

Regards,

Igor

Sent: Friday, January 10, 2020 at 3:00 AM
From: "Michael DeBellis" <[hidden email]>
To: "User support for WebProtege and Protege Desktop" <[hidden email]>
Subject: [protege-user] Enterprise Data Modeling with OWL and Protege?
I have a potential client, a large social media company, that is building an enterprise wide data model and may be interested in using OWL and Protege as part of the project. If anyone knows of any good papers or other sources on the topic please let me know. In addition to modeling the social media domain security and integrity constraints on data are two of the most significant issues for them. I plan to look at SHACL regarding the integrity constraints, I've heard it mentioned often but have never used it. If anyone has feedback on those two topics as well please let me know. thanks,
 
Michael
_______________________________________________ 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
_______________________________________________ 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

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