[OTDev] URI Lists
Nina Jeliazkova nina at acad.bgSun Feb 7 22:21:04 CET 2010
- Previous message: [OTDev] URI Lists
- Next message: [OTDev] Some things have to be clarified
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Hi, chung wrote: > Hi, > I'm also in favor of introducing new entities (classes, properties etc) > if we can do what we need using the existing ones. However, if you have > a collection of URIs you can't really say what this collection is about > even if you can describe in detail each one of the included entities. > For example, if you have a collection of algorithms in RDF you can > describe each one of the algorithm, but how can you say that this list > is called "my list" and has the description "the best algorithms we > have". One could assign each algorithm a property "best algorithm" . Or assign model a property "validated model". Then it is easy to search by sparql "give me all algorithms , which have property "best algorithm". For assigning "My List", there is already a property "creator", one could easily look for all algorithms , having creator "chung". > We provide lists of URIs in RDF at > http://opentox.ntua.gr:3000/algorithm and > http://opentox.ntua.gr:3000/model and > http://opentox.ntua.gr:3000/model?algorihtm=mlr etc... But I can't find > out a way to describe the list itself. > There is not really a need to assign metadata to the list, since all entries in the list can be described with their RDF representation. Could you please modify the services to include not only URLs, but full representation of objects (algorithm, models, etc.)? Otherwise, it will be difficult to use NTUA models in ontology service and Fastox. Thanks, Nina > On Sat, 2010-02-06 at 07:22 +0200, Nina Jeliazkova wrote: > >> chung wrote: >> >>> On Fri, 2010-02-05 at 19:45 +0200, Nina Jeliazkova wrote: >>> >>> >>>> Hi, >>>> >>>> What would be the purpose of uri-lists in RDF ? >>>> >>>> >>>> >>> For example, a client asks for a list of all QSAR models on the server. >>> Shouldn't these be available in rdf apart from text/uri-list? And what >>> >>> >> Yes, but there is no need to invent new objects/syntax, just add all >> objects from the list to a single jenaModel and serialize to RDF. >> >>> is more, how can you tell what is this list is about? >>> >> Just by sending RDF representations of objects, ot:Models in this >> case. There is no need to invent another representation - we already >> have it, >> - and all the code already written will work , >> - one would be able to import all the models and their metadata in >> Ontology service and make it available for sparql queries by simply >> posting /model URL to the Ontology service >> >> For example try >> >> (list of models) >> >> curl -X GET -H Accept:application/rdf+xml >> http://ambit.uni-plovdiv.bg:8080/ambit2/model >> >> or (list of algorithms) >> >> curl -X GET -H Accept:application/rdf+xml >> http://ambit.uni-plovdiv.bg:8080/ambit2/algorithm >> >> or (list of tasks) >> >> curl -X GET -H Accept:application/rdf+xml >> http://ambit.uni-plovdiv.bg:8080/ambit2/task >> >> or (list of datasets) >> >> curl -X GET -H Accept:application/rdf+xml >> http://ambit.uni-plovdiv.bg:8080/ambit2/dataset >> >> or (list of dependent variables pf a model) >> >> curl -X GET -H Accept:application/rdf+xml >> http://ambit.uni-plovdiv.bg:8080/ambit2/model/11/dependent >> >>> For example, you >>> want to include in the representation that this is a uri list of all >>> >>> >>> >> In RDF it should be representations of ot:Models , which of them >> identified by URI >> >> curl -X GET -H "Accept:text/n3" >> http://ambit.uni-plovdiv.bg:8080/ambit2/model?max=2 >> >> >> @prefix dc: <http://purl.org/dc/elements/1.1/> . >> @prefix ot: <http://www.opentox.org/api/1.1#> . >> @prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> . >> @prefix ota: >> <http://www.opentox.org/algorithmTypes.owl#> . >> @prefix xsd: <http://www.w3.org/2001/XMLSchema#> . >> @prefix owl: <http://www.w3.org/2002/07/owl#> . >> @prefix otee: <http://www.opentox.org/echaEndpoints.owl#> . >> @prefix rdf: >> <http://www.w3.org/1999/02/22-rdf-syntax-ns#> . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12387> >> a ot:Feature . >> >> ot:Dataset >> a owl:Class . >> >> dc:identifier >> a owl:AnnotationProperty . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/29683> >> a ot:Feature . >> >> dc:type >> a owl:AnnotationProperty . >> >> ot:Model >> a owl:Class . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/model/2> >> a ot:Model ; >> dc:creator "N/A" ; >> dc:date "N/A" ; >> dc:format "N/A" ; >> dc:identifier >> "http://ambit.uni-plovdiv.bg:8080/ambit2/model/2"^^xsd:anyURI ; >> dc:title "ToxTree: Benigni/Bossa rules for >> carcinogenicity and mutagenicity" ; >> ot:algorithm >> <http://ambit.uni-plovdiv.bg:8080/ambit2/algorithm/toxtreecarc> ; >> ot:predictedVariables >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12385> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12387> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12382> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12383> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12386> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12384> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12389> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12390> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12381> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12388> . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12382> >> a ot:Feature . >> >> ot:Algorithm >> a owl:Class . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12384> >> a ot:Feature . >> >> dc:date >> a owl:AnnotationProperty . >> >> ot:dependentVariables >> a owl:ObjectProperty . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12390> >> a ot:Feature . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/algorithm/toxtreecarc> >> a ot:Algorithm . >> >> dc:title >> a owl:AnnotationProperty . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/algorithm/pka> >> a ot:Algorithm . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12388> >> a ot:Feature . >> >> ot:predictedVariables >> a owl:ObjectProperty . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12360> >> a ot:Feature . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12386> >> a ot:Feature . >> >> dc:description >> a owl:AnnotationProperty . >> >> ot:Feature >> a owl:Class . >> >> ot:independentVariables >> a owl:ObjectProperty . >> >> dc:format >> a owl:AnnotationProperty . >> >> ot:Parameter >> a owl:Class . >> >> ot:algorithm >> a owl:ObjectProperty . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/model/1> >> a ot:Model ; >> dc:creator "N/A" ; >> dc:date "N/A" ; >> dc:format "N/A" ; >> dc:identifier >> "http://ambit.uni-plovdiv.bg:8080/ambit2/model/1"^^xsd:anyURI ; >> dc:title "pKa" ; >> ot:algorithm >> <http://ambit.uni-plovdiv.bg:8080/ambit2/algorithm/pka> ; >> ot:predictedVariables >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12360> , >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/29683> . >> >> dc:creator >> a owl:AnnotationProperty . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12385> >> a ot:Feature . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12383> >> a ot:Feature . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12389> >> a ot:Feature . >> >> <http://ambit.uni-plovdiv.bg:8080/ambit2/feature/12381> >> a ot:Feature . >> >> >>> models produced by the user 'chung' using the algorithm >>> 'http://abc.com/algorithm/eee' etc... >>> >>> >> POST top NTUA models URL into ontology service (once it supports >> RDF ) and then query the ontology service by SPARQL >> >> ?model dc:creator "chung". >> ?model ot:algorithm <http://abc.com/algorithm/eee>. >> >> >> Best regards, >> Nina >> >>> >>> >>>> Regards, >>>> Nina >>>> >>>> chung wrote: >>>> >>>> >>>>> Hi all, >>>>> >>>>> I have a little question regarding the representation of URI lists in >>>>> rdf: Is it ok to include all URIs in the RDF arranged in triples like: >>>>> {uri1, type, type1} >>>>> {uri2, type, type2} >>>>> ? >>>>> >>>>> >>>>> Following this pattern there is no way to assign a title to the List >>>>> itself or other metadata related to the uri list and what is more there >>>>> is no Ontological Entity (e.g. class) for describing such lists. >>>>> >>>>> I think there's a more structured way to do that, such as: >>>>> {uri1, type, type1} >>>>> {uri2, type, type2} >>>>> { someListUri, { >>>>> type ot:UriList >>>>> hasUri uri1 >>>>> hasUri uri2 >>>>> hasUri uri3 >>>>> title "my list of URIs" >>>>> } >>>>> } >>>>> Is there a standard way of representing URI Lists in RDF? If not, we >>>>> could introduce the class ot:URIList and the property ot:hasUri... >>>>> >>>>> Best Regards, >>>>> Pantelis >>>>> >>>>> _______________________________________________ >>>>> Development mailing list >>>>> Development at opentox.org >>>>> http://www.opentox.org/mailman/listinfo/development >>>>> >>>>> >>>>> >>>> _______________________________________________ >>>> Development mailing list >>>> Development at opentox.org >>>> http://www.opentox.org/mailman/listinfo/development >>>> >>>> >>>> >>> >>> > >
- Previous message: [OTDev] URI Lists
- Next message: [OTDev] Some things have to be clarified
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Development mailing list