thank you Lorenz for your time and guidance. I am working on it and will get back to you, if required.
Regards On Sun, Nov 12, 2017 at 4:41 PM, Lorenz Buehmann < [email protected]> wrote: > > > On 12.11.2017 11:16, Sidra shah wrote: > > Hello Lorenz. thank you again. > > > > Yes I want to " load the DBpedia data into my ontology". > > > > Actually my ontology data and the data I need from Dbpedia are so related > > that I want Dbpedia data to be imported/stored in my ontology. > > > > According to my instructor, use any way but the "extraction of Dbpedia > > data" should be via ontology. It means do not use Dbpedia endpoint inside > > your application. > As I said: > 1) use SPARQL CONSTRUCT to get the necessary data for your application > form the DBpedia endpoint > 2) add this data to your ontology > 3) then do whatever you're doing in your application based on that local > data. > > > > Regards > > > > On Sat, Nov 11, 2017 at 7:17 PM, Lorenz Buehmann < > > [email protected]> wrote: > > > >> > >> On 11.11.2017 14:58, Sidra shah wrote: > >>> Thank you again Lorenz, > >>> > >>> I have (partially) got your point and quite useful explanation here. > >>> > >>> All my (required) data is from Dbpedia. > >>> > >>> My point is to get this data using Semantic web application (using > Jena ) > >>> via SPARQL query. > >> I still don't understand what your "application" does...accessing the > >> data means to either load the DBpedia data into your ontology in advance > >> or to use federated SPARQL queries at application runtime. > >>> One way is to directly use Dbpedia endpoint (which can not be re-used > and > >>> shared) > >> Why not? The whole DBpedia dataset is open data and can be downloaded > >> and used by everyone locally. > >>> I want to use the other way in which I access Dbpedia data using my > >>> ontology. I have read an article on stackoverflow which says > >>> "include the link of required Dbpedia resource in the OWL > NamedIndividual > >>> tab like http:dbpedia/resource/name and query it inside Semantic Web > >>> application (via SPARQL) as you query local data of ontology". > >> If your ontology just contains DBpedia resources, then it's nothing more > >> than a subset of the DBpedia dataset. What prevents from retrieving the > >> data you really need in your application and then use that data in your > >> Web application locally? > >> > >> In general, reusing IRI as OWL individual in your local ontology is > >> similar to just "talking" about the same individual in your ontology. > >> That doesn't necessary mean that you're having access to data about this > >> individual provided by others like the people who extracted the DBpedia > >> resource from Wikipedia infoboxes. > >>> But you earlier mentioned that one need CONSTRUCT queries in this > >>> situation. I am not sure how and why COSTRUCT can be used be used in > that > >>> case. > >> In your initial question you said that you want to extract "some > >> triples". You're the only person who know which triples, thus, you're > >> the one that should be able to query for this data. And I mentioned > >> SPARQL CONSTRUCT because this type of query returns a set of RDF triples > >> compared to SPARQL SELECT queries which returns a resultset. > >>> My instructor recommend me to use Dbpedia knowledge using Ontologies. > >>> > >>> Regards > >>> > >>> <https://www.avast.com/sig-email?utm_medium=email&utm_ > >> source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> > >>> Virus-free. > >>> www.avast.com > >>> <https://www.avast.com/sig-email?utm_medium=email&utm_ > >> source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link> > >>> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2> > >>> > >>> On Sat, Nov 11, 2017 at 3:39 PM, Lorenz Buehmann < > >>> [email protected]> wrote: > >>> > >>>> On 11.11.2017 12:55, Sidra shah wrote: > >>>>> Hello Lorenz and thank you for your information. > >>>>> > >>>>> * All the data of this resource is still located in the DBpedia > dataset > >>>>> > >>>>> If it is the case, then why we provide links to Dbpedia resource > inside > >>>>> Protege editor? All I want to re-use the data/information of Dbpedia > >>>>> > >>>>> Its then better that we use rdfs:seeAlso and provide Dbpedia > resource, > >>>> like > >>>>> www.myOntology.org/Oxford and then use > >>>>> rdfs:seeAlso http:dbpedia.org/resource/Oxford > >>>> I guess you're mixing up things here. Indeed it's fine to reuse > resource > >>>> from the Web of Data. I mean, that's in general how Linked Data is > >>>> supposed to work. Ok, sometimes it'S also recommended to define your > own > >>>> resources and relate those to external datasets via owl:sameAs, but > the > >>>> result is more or less the same. The data is located at different > >> places. > >>>> But, and that's what you have to understand: whatever you're doing > with > >>>> your local data, querying, inferencing, etc. - the tool/framework/API > >>>> you're using for that has to be able to retrieve the data from > different > >>>> locations if the data is physically located at different locations. > >>>> Seems quite obvious or not? > >>>> > >>>> And now it's up to you: given that you're reusing DBpedia resources in > >>>> your ontology: > >>>> 1) how does the SPARQL query engine working on your local ontology > know > >>>> where the data comes from? Again, it should be quite obvious that it's > >>>> up to you to do all the setup, which brings us to the concept of > >>>> federated query processing... > >>>>> Regards > >>>>> > >>>>> > >>>>> > >>>>> <https://www.avast.com/sig-email?utm_medium=email&utm_ > >>>> source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=icon> > >>>>> Virus-free. > >>>>> www.avast.com > >>>>> <https://www.avast.com/sig-email?utm_medium=email&utm_ > >>>> source=link&utm_campaign=sig-email&utm_content=webmail&utm_term=link> > >>>>> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2> > >>>>> > >>>>> On Sat, Nov 11, 2017 at 1:25 PM, Lorenz Buehmann < > >>>>> [email protected]> wrote: > >>>>> > >>>>>>> The second point is if we enter individual in Protege (Create New > OWL > >>>>>>> Individual) and then enter URI like "http:dbpedia.org/resource". > >>>>>> I understand. But what do you expect to be happened with this > step?All > >>>>>> that you did is to create an OWL individual with the URI of the > >> DBpedia > >>>>>> resource. All the data of this resource is still located in the > >> DBpedia > >>>>>> dataset which is > >>>>>> a) available via RDF dumps or > >>>>>> b) the public DBpedia SPARQL endpoint > >>>>>> c) HTTP GET request according to the Linked Data priciple > >>>>>> > >>>>>> But the data is **not** in your local ontology and neither Protege > nor > >>>>>> the built-in SPARQL plugin would have access to it. > >>>>>>> By better I mean better in general (performance, re-use).etc. Will > it > >>>> be > >>>>>>> considered a "Dbpedia resource" if we just include its URI in > Protege > >>>>>>> editor and then query it locally like we query traditional data in > >>>>>> Protege > >>>>>>> (Ontology). > >>>>>> How do you query it locally? Which API, which triple store, etc? > >>>>>> > >>>>>> In general, what is the use-case? > >>>>>> > >>>>>>> <https://www.avast.com/sig-email?utm_medium=email&utm_ > >>>>>> source=link&utm_campaign=sig-email&utm_content=webmail&utm_ > term=icon> > >>>>>>> Virus-free. > >>>>>>> www.avast.com > >>>>>>> <https://www.avast.com/sig-email?utm_medium=email&utm_ > >>>>>> source=link&utm_campaign=sig-email&utm_content=webmail&utm_ > term=link> > >>>>>>> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2> > >>>>>>> > >>>>>>> On Sat, Nov 11, 2017 at 1:05 PM, Lorenz Buehmann < > >>>>>>> [email protected]> wrote: > >>>>>>> > >>>>>>>> 1. Define "better" > >>>>>>>> > >>>>>>>> 2. I don't understand what you mean by the second point ... what > is > >> an > >>>>>>>> "IRI editor"??? And then, how would that extract "some triples"? > >>>>>>>> > >>>>>>>> > >>>>>>>> As I don't know what you're asking about and to keep it short, the > >>>>>>>> common way to extract RDF triples from and RDF dataset is to use a > >>>>>>>> SPARQL CONSTRUCT query that matches those "some triples". > >>>>>>>> > >>>>>>>> > >>>>>>>> On 10.11.2017 17:11, Sidra shah wrote: > >>>>>>>>> Hello > >>>>>>>>> > >>>>>>>>> For instance, if we have to get some triples from Dbpedia, which > >> one > >>>> is > >>>>>>>>> better way to get? > >>>>>>>>> > >>>>>>>>> (1) Directly use Dbpedia endpoint inside application? > >>>>>>>>> > >>>>>>>>> (2) Use Ontology and use IRI editor like > >>>>>> dbpedia.org/resource/SOMETHING? > >>>>>>>>> Thank you > >>>>>>>>> > >> > >
