I'm trying to write a SPARQL request, where I want to query country names from my own dataset, use the dbpedia resource to retrieve the countrie's latitude and longitude and then calculate the distance among them with the bif:st_distance function. The latitude and longitude I want to use are floats, so I have to convert them to points with bif:st_point first, since the function takes points only.
On my local machine I'm running a apache jena-fuseki server with my dataset loaded. I tried the following:
PREFIX rate: <http://MYNAME.bplaced.net/rating/index.ttl#>
PREFIX dbr: <http://dbpedia.org/resource/>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
PREFIX dbo: <http://dbpedia.org/ontology/>
PREFIX bif: <bif:>
PREFIX dbr: <http://dbpedia.org/resource/>
PREFIX geo: <http://www.w3.org/2003/01/geo/wgs84_pos#>
PREFIX afn: <http://jena.apache.org/ARQ/function#>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
SELECT ?from ?to ?latituded ?longituded ?latitudeo ?longitudeo ?distance WHERE{
{SELECT ?from ?to (AVG(?latd) as ?latituded)
(AVG(?longd) as ?longituded)
(AVG(?lato) as ?latitudeo)
(AVG(?longo) as ?longitudeo) WHERE {
?rating rate:from dbr:Germany.
?rating rate:year "1965"^^xsd:gYear.
?rating rate:from ?from.
?rating rate:to ?to.
SERVICE <http://dbpedia.org/sparql>{
?from geo:lat ?latd.
?from geo:long ?longd.
?to geo:lat ?lato.
?to geo:long ?longo.
}
} GROUP BY ?from ?to
}
BIND(bif:st_distance(
bif:st_point(?latituded,?longituded),
bif:st_point(?latitudeo,?longitudeo)
)AS ?distance).
}
This leaves me with this result:
The query result
As you can see: The distance column remains empty. I am sure it has something to do with whether or not the BIND statement is located within the SERVICE. Might some values be invisible to the BIND?
What am I doing wrong or what can I do differently to solve this?
Thanks in advance!
The spatial functions in the bif namespace are specific to Virtuoso, the software used by DBpedia. Therefore they will only work in the SERVICE clause where the query is executed on the Virtuoso server. So either use these functions in the SERVICE clause or take a look at the Jena spatial functions.
Related
I have a list of artstyles in graphDB, i am trying to use the SERVICE function to get their labels from Wikidata with this query:
PREFIX gp: <http://www.semanticweb.org/kandd/group76/final_project#>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
SELECT ?movement ?label
WHERE{
?artist gp:hasArtStyle ?movement.
SERVICE <https://query.wikidata.org/sparql>{
?movement rdfs:label ?label .
FILTER (langMatches( lang(?label), "EN" ) )
}
}
note that gp is a namespace that only exists in my graph, not anywhere on the internet and also note that ?movement contains a list of valid Wikidata URIs such as http://www.wikidata.org/entity/Q186030
yet still the response I get is:
Error 500: error
Query evaluation error: org.eclipse.rdf4j.query.QueryEvaluationException: org.eclipse.rdf4j.query.QueryEvaluationException: java.io.IOException: Unkown record type: 83 (HTTP status 500)
What am I doing wrong?
Remember that you query is handled from the inside to the outside, meaning that the service part is handled first, and then the part where you use your own specific property.
Currently, your query on WikiData is very general. You ask for everything that has a rdfs:label, and then filter on all the English labels it returns.
Given this, my guess is that you query simply times out. Instead, I would try something like this:
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX wdt: <http://www.wikidata.org/prop/direct/>
PREFIX wd: <http://www.wikidata.org/entity/>
SELECT *
WHERE{
SERVICE <https://query.wikidata.org/sparql>{
?artist wdt:P101 wd:Q186030 ; #Field of Work is contemporary art
wdt:P31 wd:Q5 ; #instance of Human
rdfs:label ?name . #get the label
FILTER (langmatches(lang(?name), "en"))
}
}
If I try this in GraphDB, it returns 156 results.
I have a Graph Database that implements only one side of relationships in the model structures. For instance, All Broader relationships are mapped as skos:broader. All Narrower relationships are mapped as ^skos:broader (for whatever reasons they chose to do this, I have no control over this.)
I am trying to write a SPARQL Query that delivers something like
Query for all relationships of a given IRI, and their associated values.
The goal is to make a JSON or XML Representation of all the content.
Here is what I have working.
PREFIX skosxl: <http://www.w3.org/2008/05/skos-xl#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX owl: <http://www.w3.org/2002/07/owl#>
PREFIX skos: <http://www.w3.org/2004/02/skos/core#>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
select ?relation ?value
where
{
<CONCEPT_URI> ?r ?c
OPTIONAL {
?r rdfs:label|skosxl:prefLabel ?l
}
OPTIONAL {
?c skosxl:prefLabel/skosxl:literalForm|skosxl:literalForm ?d .
}
BIND ( if (bound(?d), ?d, ?c) as ?value )
BIND ( if (bound(?l), ?l, strafter(?r, '#')) as ?relation )
}
This works to give me all the concept relationships (IE: Broader), but it does not give me the narrower items (IE: ^broader).
Is there anyway that I can extend this generic query to produce the inverses of any relationships it finds, as well? I Tried simply adding the ^ to the ?r in the where clause but it does not like that syntactically.
Thanks,
Hi Below is my sparql query which I am executing on https://dbpedia.org/sparql. However I am getting an error "Invalid floating point value converting 'United States:'" because of the presence of some characters values in grossincome and If I change my query to str(?grossincome) as ?grossincome1) then it executes correctly. My question is how can I convert all currencies on same scale to avoid this error?
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX dcterms: <http://purl.org/dc/terms/>
PREFIX dbo: <http://dbpedia.org/ontology/>
PREFIX dbp: <http://dbpedia.org/property/>
PREFIX foaf: <http://xmlns.com/foaf/0.1/>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
select distinct (str(?resource) as ?movietitle) (xsd:float(?budget) as ?budget1) (xsd:float(?grossincome) as ?grossincome1) (str(?Country) as ?country1) ?ReleaseDate where {
?movie foaf:name ?resource.
?movie dbo:budget ?budget.
?movie dbp:gross ?grossincome.
?movie dbp:country ?Country.
?movie dbo:releaseDate ?ReleaseDate.
FILTER (lang(?resource) = 'en').
}
The data in DBpedia is not perfect as you already recognized it's quite heterogeneous for the untyped property dbp:gross. Among others, it contains values like BDT 2.52 crores, $34,994,648 USD, AU$178,000, 6892098.0, so it is almost impossible to handle different currencies via a SPARQL query such that it returns the values in the same currency.
I mean, you would have to know all different currencies and formats that occur and then handle each separately. But that's more the task of application logic and not the query itself.
Alternatively, you could try to use the data property http://dbpedia.org/ontology/gross instead, which is supposed to relate to values in $. At least that's what its label "gross ($)" states.
I have this query
PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX foaf: <http://xmlns.com/foaf/0.1/>
PREFIX dbpedia: <http://dbpedia.org/>
PREFIX dbpedia_property: <http://dbpedia.org/property/>
PREFIX dbpedia_ontology: <http://dbpedia.org/ontology/>
PREFIX yago: <http://dbpedia.org/class/yago/>
PREFIX schema: <http://schema.org/>
SELECT * WHERE
{
{
SELECT ?school
WHERE
{
?school rdf:type yago:EducationalInstitution108276342 .
FILTER ( contains(str(?school), "Australia") )
}
ORDER BY ?school
}
}
Don't mind the extra brackets as this is part of a larger query.
What I want to know is why thi http://dbpedia.org/page/Academic_structure_of_the_Australian_National_University is included in the results since I specify rdf:type yago:EducationalInstitution108276342. This property is not included in the resource page. I'm using this endpoint: http://dbpedia.org/sparql
Looks like a bug in the Pubby Web interface or in the query that is used to get the data that will be shown.
The query
SELECT * WHERE{
<http://dbpedia.org/resource/Academic_Structure_of_the_Australian_National_University> ?p ?o
}
returns the necessary rdf:type statement.
The other strange thing is that even a SPARQL DESCRIBE query does not return the rdd:type triples:
DESCRIBE <http://dbpedia.org/resource/Academic_Structure_of_the_Australian_National_University>
Although DESCIBE is not really defined in the specs, a user would expect those triples for sure. And maybe this kind of query is used to retrieve the data for the Web pages of resources.
Why does this SPARQL query return no data?
PREFIX dbpedia-owl: <http://dbpedia.org/ontology/>
PREFIX dbpedia: <http://dbpedia.org/resource/>
SELECT *
WHERE {
<http://dbpedia.org/resource/Louis,_Prince_of_Condé_(1530–1569)> dbpedia-owl:abstract ?abstract
}
LIMIT 1
If you look at the DBpedia page, it shows the person has an abstract. Is it to do with the brackets in the URL? If so, how can I get round this?
This URI does not lead to the same result as the DBpedia page - for what ever reason. You can see this with
PREFIX dbpedia-owl: <http://dbpedia.org/ontology/>
PREFIX dbpedia: <http://dbpedia.org/resource/>
SELECT *
WHERE {
<http://dbpedia.org/resource/Louis,_Prince_of_Condé_(1530–1569)> ?p ?o
}
LIMIT 100
But it has an owl:sameAs relation to
http://dbpedia.org/resource/Louis,_Prince_of_Cond%C3%A9_(1530%E2%80%931569)
That means if you use this URI in your query, it should work as expected. But you should indeed apply a FILTER on the language, e.g. 'en' for English abstracts.
As AKSW mentions, the resource actually doesn't have many properties, but is connected to the "canonical" version by an owl:sameAs link. You can keep using the IRI that you're using now, follow owl:sameAs in either direction to any of its equal resources (let's call them ?s), and then ask for the abstract of ?s. (And then it's not a bad idea to filter by language, if that's applicable.) You can do this with a query like this (note that the current DBpedia endpoint uses dbo:, now, not the older dbpedia-owl:):
select ?abstract where {
<http://dbpedia.org/resource/Louis,_Prince_of_Condé_(1530–1569)> (owl:sameAs|^owl:sameAs)* ?s .
?s dbo:abstract ?abstract .
filter langMatches(lang(?abstract),'en')
}
It does not have dbpedia-owl:abstract predicate. If you list its predicates you find the following properties:
http://www.w3.org/2002/07/owl#sameAs
http://xmlns.com/foaf/0.1/name
http://purl.org/dc/elements/1.1/description
http://dbpedia.org/ontology/alias
http://dbpedia.org/ontology/birthYear
http://dbpedia.org/ontology/deathYear
http://dbpedia.org/ontology/viafId
http://dbpedia.org/ontology/deathPlace
http://dbpedia.org/ontology/deathDate
http://dbpedia.org/ontology/birthPlace
http://dbpedia.org/ontology/birthDate