OWL APIs, Triple Stores, OWL DL Rasoners

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

OWL APIs, Triple Stores, OWL DL Rasoners

Nick Khamis
Hello Everyone,

We are looking for a solution that would allow us to transfer our OWL2
knowledge base to a full fledged application.
What we are looking for are:

* An OWL API capable of creating OWL 2 models: The Protege API is
great however comes with much to much overhead,
i.e., the editor related implemanation. Is the Jena api capable of
creating OWL 2 ontology models? Are there any new,
and active ontology APIs currently out there?

* A triple store capable of storing OWL 2 expressions, and performing
deduction using the widely used reasoners. It is
important for us to support OWL or OWL 2 reasoning on the triple store
level. Much like Sesame does with OWLIM.
The triple store must also be able to inteface with the Pellet reasoner.

* Interface between the reasoners and the triple stores. I would
assume at this early stage that deduction on the triple
store level using a pellet would be done through an interface of some
sort (e.g., Triple Store -> Jena -> Pellet). If I am
not mistaken, what are some of the soultions used out there for this.

I have looked into projects such as PelletDB and Allegrograph which
look great however, we are looking for actively
developed open source solutions

Thanks in Advnace,

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

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

Re: OWL APIs, Triple Stores, OWL DL Rasoners

Matthew Horridge-2
Administrator
Hi,

> We are looking for a solution that would allow us to transfer our OWL2
> knowledge base to a full fledged application.
> What we are looking for are:
>
> * An OWL API capable of creating OWL 2 models: The Protege API is
> great however comes with much to much overhead,
> i.e., the editor related implemanation. Is the Jena api capable of
> creating OWL 2 ontology models? Are there any new,
> and active ontology APIs currently out there?

There's the OWL API  (http://owlapi.sourceforge.net)

Pellet's native Ortiz API (http://weblog.clarkparsia.com/2010/12/17/ortiz-pellet/)

Any RDF API can work at the level triples that are used to encode OWL 2 objects.  However, this might be too low level for you.

> * A triple store capable of storing OWL 2 expressions, and performing
> deduction using the widely used reasoners. It is
> important for us to support OWL or OWL 2 reasoning on the triple store
> level. Much like Sesame does with OWLIM.
> The triple store must also be able to inteface with the Pellet reasoner.

Take a look at Clark and Parsia's StarDog software (http://stardog.com/).  It looks very comprehensive and Pellet is a key component.

> * Interface between the reasoners and the triple stores. I would
> assume at this early stage that deduction on the triple
> store level using a pellet would be done through an interface of some
> sort (e.g., Triple Store -> Jena -> Pellet). If I am
> not mistaken, what are some of the soultions used out there for this.

See above.

Cheers,

Matthew


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

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

Re: OWL APIs, Triple Stores, OWL DL Rasoners

Nick Khamis
Hello Matthew,

Thank you so much for your response. This is all amazing information!
Could you recomment a low level RDF API. Reason being, they tend
to be more efficent/faster. The benefits could far outweigh the drawbacks,
as long as there is already a nice implementation of the owl2 object wrappers?

Thanks in Advance,

Ninus.

On Wed, Nov 9, 2011 at 7:57 PM, Matthew Horridge
<[hidden email]> wrote:

> Hi,
>
>> We are looking for a solution that would allow us to transfer our OWL2
>> knowledge base to a full fledged application.
>> What we are looking for are:
>>
>> * An OWL API capable of creating OWL 2 models: The Protege API is
>> great however comes with much to much overhead,
>> i.e., the editor related implemanation. Is the Jena api capable of
>> creating OWL 2 ontology models? Are there any new,
>> and active ontology APIs currently out there?
>
> There's the OWL API  (http://owlapi.sourceforge.net)
>
> Pellet's native Ortiz API (http://weblog.clarkparsia.com/2010/12/17/ortiz-pellet/)
>
> Any RDF API can work at the level triples that are used to encode OWL 2 objects.  However, this might be too low level for you.
>
>> * A triple store capable of storing OWL 2 expressions, and performing
>> deduction using the widely used reasoners. It is
>> important for us to support OWL or OWL 2 reasoning on the triple store
>> level. Much like Sesame does with OWLIM.
>> The triple store must also be able to inteface with the Pellet reasoner.
>
> Take a look at Clark and Parsia's StarDog software (http://stardog.com/).  It looks very comprehensive and Pellet is a key component.
>
>> * Interface between the reasoners and the triple stores. I would
>> assume at this early stage that deduction on the triple
>> store level using a pellet would be done through an interface of some
>> sort (e.g., Triple Store -> Jena -> Pellet). If I am
>> not mistaken, what are some of the soultions used out there for this.
>
> See above.
>
> Cheers,
>
> Matthew
>
>
> _______________________________________________
> protege-discussion mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>
> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
>
_______________________________________________
protege-discussion mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-discussion

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

Re: OWL APIs, Triple Stores, OWL DL Rasoners

Timothy Redmond
On 11/09/2011 05:17 PM, Nick Khamis wrote:
> Hello Matthew,
>
> Thank you so much for your response. This is all amazing information!
> Could you recomment a low level RDF API. Reason being, they tend
> to be more efficent/faster. The benefits could far outweigh the drawbacks,
> as long as there is already a nice implementation of the owl2 object wrappers?

Try the Sesame api (www.openrdf.org).  Seems very nice.  I don't know if
it works with pellet though.  Jena may also have some advantages.

> Reason being, they tend
> to be more efficent/faster.

Are you sure about this?  There are often claims of scalability, which
have some truth, but need to be carefully understood.  You need to make
sure that you are making a comparison of like things and that you are
getting the performance in the area that you need.  For example, I
suspect that there are no rdf based tools that could do a classification
of snomed in a few seconds.  (You did mention inference.)  You could
probably find other examples where triple stores work best.

-Timothy




> The benefits could far outweigh the drawbacks,
> as long as there is already a nice implementation of the owl2 object wrappers?


>
> Thanks in Advance,
>
> Ninus.
>
> On Wed, Nov 9, 2011 at 7:57 PM, Matthew Horridge
> <[hidden email]>  wrote:
>> Hi,
>>
>>> We are looking for a solution that would allow us to transfer our OWL2
>>> knowledge base to a full fledged application.
>>> What we are looking for are:
>>>
>>> * An OWL API capable of creating OWL 2 models: The Protege API is
>>> great however comes with much to much overhead,
>>> i.e., the editor related implemanation. Is the Jena api capable of
>>> creating OWL 2 ontology models? Are there any new,
>>> and active ontology APIs currently out there?
>> There's the OWL API  (http://owlapi.sourceforge.net)
>>
>> Pellet's native Ortiz API (http://weblog.clarkparsia.com/2010/12/17/ortiz-pellet/)
>>
>> Any RDF API can work at the level triples that are used to encode OWL 2 objects.  However, this might be too low level for you.
>>
>>> * A triple store capable of storing OWL 2 expressions, and performing
>>> deduction using the widely used reasoners. It is
>>> important for us to support OWL or OWL 2 reasoning on the triple store
>>> level. Much like Sesame does with OWLIM.
>>> The triple store must also be able to inteface with the Pellet reasoner.
>> Take a look at Clark and Parsia's StarDog software (http://stardog.com/).  It looks very comprehensive and Pellet is a key component.
>>
>>> * Interface between the reasoners and the triple stores. I would
>>> assume at this early stage that deduction on the triple
>>> store level using a pellet would be done through an interface of some
>>> sort (e.g., Triple Store ->  Jena ->  Pellet). If I am
>>> not mistaken, what are some of the soultions used out there for this.
>> See above.
>>
>> Cheers,
>>
>> Matthew
>>
>>
>> _______________________________________________
>> protege-discussion mailing list
>> [hidden email]
>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>
>> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03
>>
> _______________________________________________
> protege-discussion mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>
> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03

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

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

Re: OWL APIs, Triple Stores, OWL DL Rasoners

Nick Khamis
Hello Timothy,

Thank you so much for your response. When you refer to the sesame api,
you are referring
to the triple store correct? Do you know if they provide an api for
ontology model creation
such as The OWLAPI or Ortiz?

> For example, I suspect that there are no rdf based tools that could do a classification of
> snomed in a few seconds.

Before doing any type of classification our focus would first be:

Find an API that can creates OWL/2 models, such as Ortiz, Jena, The OWL API.
We would also consider low level rdf api if there are performance
benefits, but of course we would
need to create the wrappers to support owl 2 expressions. As mentioned
by Matthew earlier.
My first question would be, what low level rdf api with existing owl
wrappers are currently being used
to create OwlModels (Jena?).

Once we have the ontology models created, then we would consider
triple stores, this is rather simple,
it could be Virtuso, Sesame etc.. The issue arise when trying to apply
reasoning services on the stored
triples. Namely, can the triple store interface directly with the
owl/2 reasoners, or inderctly using Jena etc..
The triple store should also be able to manage multiple owl models
within a given context enabling things
such as owl:import and prefix resolution.....

Thanks in Advance,

Nick.


On Fri, Nov 11, 2011 at 11:28 AM, Timothy Redmond <[hidden email]> wrote:

> On 11/09/2011 05:17 PM, Nick Khamis wrote:
>>
>> Hello Matthew,
>>
>> Thank you so much for your response. This is all amazing information!
>> Could you recomment a low level RDF API. Reason being, they tend
>> to be more efficent/faster. The benefits could far outweigh the drawbacks,
>> as long as there is already a nice implementation of the owl2 object
>> wrappers?
>
> Try the Sesame api (www.openrdf.org).  Seems very nice.  I don't know if it
> works with pellet though.  Jena may also have some advantages.
>
>> Reason being, they tend
>> to be more efficent/faster.
>
> Are you sure about this?  There are often claims of scalability, which have
> some truth, but need to be carefully understood.  You need to make sure that
> you are making a comparison of like things and that you are getting the
> performance in the area that you need.  For example, I suspect that there
> are no rdf based tools that could do a classification of snomed in a few
> seconds.  (You did mention inference.)  You could probably find other
> examples where triple stores work best.
>
> -Timothy
>
>
>
>
>> The benefits could far outweigh the drawbacks,
>> as long as there is already a nice implementation of the owl2 object
>> wrappers?
>
>
>>
>> Thanks in Advance,
>>
>> Ninus.
>>
>> On Wed, Nov 9, 2011 at 7:57 PM, Matthew Horridge
>> <[hidden email]>  wrote:
>>>
>>> Hi,
>>>
>>>> We are looking for a solution that would allow us to transfer our OWL2
>>>> knowledge base to a full fledged application.
>>>> What we are looking for are:
>>>>
>>>> * An OWL API capable of creating OWL 2 models: The Protege API is
>>>> great however comes with much to much overhead,
>>>> i.e., the editor related implemanation. Is the Jena api capable of
>>>> creating OWL 2 ontology models? Are there any new,
>>>> and active ontology APIs currently out there?
>>>
>>> There's the OWL API  (http://owlapi.sourceforge.net)
>>>
>>> Pellet's native Ortiz API
>>> (http://weblog.clarkparsia.com/2010/12/17/ortiz-pellet/)
>>>
>>> Any RDF API can work at the level triples that are used to encode OWL 2
>>> objects.  However, this might be too low level for you.
>>>
>>>> * A triple store capable of storing OWL 2 expressions, and performing
>>>> deduction using the widely used reasoners. It is
>>>> important for us to support OWL or OWL 2 reasoning on the triple store
>>>> level. Much like Sesame does with OWLIM.
>>>> The triple store must also be able to inteface with the Pellet reasoner.
>>>
>>> Take a look at Clark and Parsia's StarDog software (http://stardog.com/).
>>>  It looks very comprehensive and Pellet is a key component.
>>>
>>>> * Interface between the reasoners and the triple stores. I would
>>>> assume at this early stage that deduction on the triple
>>>> store level using a pellet would be done through an interface of some
>>>> sort (e.g., Triple Store ->  Jena ->  Pellet). If I am
>>>> not mistaken, what are some of the soultions used out there for this.
>>>
>>> See above.
>>>
>>> Cheers,
>>>
>>> Matthew
>>>
>>>
>>> _______________________________________________
>>> protege-discussion mailing list
>>> [hidden email]
>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>
>>> Instructions for unsubscribing:
>>> http://protege.stanford.edu/doc/faq.html#01a.03
>>>
>> _______________________________________________
>> protege-discussion mailing list
>> [hidden email]
>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>
>> Instructions for unsubscribing:
>> http://protege.stanford.edu/doc/faq.html#01a.03
>
> _______________________________________________
> protege-discussion mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>
> Instructions for unsubscribing:
> http://protege.stanford.edu/doc/faq.html#01a.03
>
_______________________________________________
protege-discussion mailing list
[hidden email]
https://mailman.stanford.edu/mailman/listinfo/protege-discussion

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

Re: OWL APIs, Triple Stores, OWL DL Rasoners

Timothy Redmond

> Once we have the ontology models created, then we would consider
> triple stores, this is rather simple,
> it could be Virtuso, Sesame etc.. The issue arise when trying to apply
> reasoning services on the stored
> triples. Namely, can the triple store interface directly with the
> owl/2 reasoners, or inderctly using Jena etc..
> The triple store should also be able to manage multiple owl models
> within a given context enabling things
> such as owl:import and prefix resolution.....

The triple store inference may be your bottleneck constraint.  I know
several reasoners that work very well with OWL and with different pros
and cons.  Mostly they don't work with triple stores.

I also know that there are rdf reasoners but they do very limited types
of inference.  If they do some OWL style inference they usually use a
proprietary version of RDF instead of a profile of OWL.  Personally this
seems like the wrong approach.

Finally there is some ongoing work on OWL inference for specific
profiles where the individual data can be retrieved from some sort of
database or conceivably a triple store.  I think that this third option
is the right answer but I am not sure how many and what tools are out
there. This is why we pointed you to the pellet solutions (Stardog).   I
know that there is other work in this direction (e.g. I am looking at
OBDA - http://obda.inf.unibz.it/protege-plugin/) but I am not sure what
the state of the art is.

The solution that you find for this may limit or define your approach
for everything else.

-Timothy


On 11/11/2011 09:17 AM, Nick Khamis wrote:

> Hello Timothy,
>
> Thank you so much for your response. When you refer to the sesame api,
> you are referring
> to the triple store correct? Do you know if they provide an api for
> ontology model creation
> such as The OWLAPI or Ortiz?
>
>> For example, I suspect that there are no rdf based tools that could do a classification of
>> snomed in a few seconds.
> Before doing any type of classification our focus would first be:
>
> Find an API that can creates OWL/2 models, such as Ortiz, Jena, The OWL API.
> We would also consider low level rdf api if there are performance
> benefits, but of course we would
> need to create the wrappers to support owl 2 expressions. As mentioned
> by Matthew earlier.
> My first question would be, what low level rdf api with existing owl
> wrappers are currently being used
> to create OwlModels (Jena?).
>
> Once we have the ontology models created, then we would consider
> triple stores, this is rather simple,
> it could be Virtuso, Sesame etc.. The issue arise when trying to apply
> reasoning services on the stored
> triples. Namely, can the triple store interface directly with the
> owl/2 reasoners, or inderctly using Jena etc..
> The triple store should also be able to manage multiple owl models
> within a given context enabling things
> such as owl:import and prefix resolution.....
>
> Thanks in Advance,
>
> Nick.
>
>
> On Fri, Nov 11, 2011 at 11:28 AM, Timothy Redmond<[hidden email]>  wrote:
>> On 11/09/2011 05:17 PM, Nick Khamis wrote:
>>> Hello Matthew,
>>>
>>> Thank you so much for your response. This is all amazing information!
>>> Could you recomment a low level RDF API. Reason being, they tend
>>> to be more efficent/faster. The benefits could far outweigh the drawbacks,
>>> as long as there is already a nice implementation of the owl2 object
>>> wrappers?
>> Try the Sesame api (www.openrdf.org).  Seems very nice.  I don't know if it
>> works with pellet though.  Jena may also have some advantages.
>>
>>> Reason being, they tend
>>> to be more efficent/faster.
>> Are you sure about this?  There are often claims of scalability, which have
>> some truth, but need to be carefully understood.  You need to make sure that
>> you are making a comparison of like things and that you are getting the
>> performance in the area that you need.  For example, I suspect that there
>> are no rdf based tools that could do a classification of snomed in a few
>> seconds.  (You did mention inference.)  You could probably find other
>> examples where triple stores work best.
>>
>> -Timothy
>>
>>
>>
>>
>>> The benefits could far outweigh the drawbacks,
>>> as long as there is already a nice implementation of the owl2 object
>>> wrappers?
>>
>>> Thanks in Advance,
>>>
>>> Ninus.
>>>
>>> On Wed, Nov 9, 2011 at 7:57 PM, Matthew Horridge
>>> <[hidden email]>    wrote:
>>>> Hi,
>>>>
>>>>> We are looking for a solution that would allow us to transfer our OWL2
>>>>> knowledge base to a full fledged application.
>>>>> What we are looking for are:
>>>>>
>>>>> * An OWL API capable of creating OWL 2 models: The Protege API is
>>>>> great however comes with much to much overhead,
>>>>> i.e., the editor related implemanation. Is the Jena api capable of
>>>>> creating OWL 2 ontology models? Are there any new,
>>>>> and active ontology APIs currently out there?
>>>> There's the OWL API  (http://owlapi.sourceforge.net)
>>>>
>>>> Pellet's native Ortiz API
>>>> (http://weblog.clarkparsia.com/2010/12/17/ortiz-pellet/)
>>>>
>>>> Any RDF API can work at the level triples that are used to encode OWL 2
>>>> objects.  However, this might be too low level for you.
>>>>
>>>>> * A triple store capable of storing OWL 2 expressions, and performing
>>>>> deduction using the widely used reasoners. It is
>>>>> important for us to support OWL or OWL 2 reasoning on the triple store
>>>>> level. Much like Sesame does with OWLIM.
>>>>> The triple store must also be able to inteface with the Pellet reasoner.
>>>> Take a look at Clark and Parsia's StarDog software (http://stardog.com/).
>>>>   It looks very comprehensive and Pellet is a key component.
>>>>
>>>>> * Interface between the reasoners and the triple stores. I would
>>>>> assume at this early stage that deduction on the triple
>>>>> store level using a pellet would be done through an interface of some
>>>>> sort (e.g., Triple Store ->    Jena ->    Pellet). If I am
>>>>> not mistaken, what are some of the soultions used out there for this.
>>>> See above.
>>>>
>>>> Cheers,
>>>>
>>>> Matthew
>>>>
>>>>
>>>> _______________________________________________
>>>> protege-discussion mailing list
>>>> [hidden email]
>>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>>
>>>> Instructions for unsubscribing:
>>>> http://protege.stanford.edu/doc/faq.html#01a.03
>>>>
>>> _______________________________________________
>>> protege-discussion mailing list
>>> [hidden email]
>>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>>
>>> Instructions for unsubscribing:
>>> http://protege.stanford.edu/doc/faq.html#01a.03
>> _______________________________________________
>> protege-discussion mailing list
>> [hidden email]
>> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>>
>> Instructions for unsubscribing:
>> http://protege.stanford.edu/doc/faq.html#01a.03
>>
> _______________________________________________
> protege-discussion mailing list
> [hidden email]
> https://mailman.stanford.edu/mailman/listinfo/protege-discussion
>
> Instructions for unsubscribing: http://protege.stanford.edu/doc/faq.html#01a.03

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

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