SPARQL Intro: A query language for RDF Alexandra Cristea 1
pronounced "sparkle“ recursive acronym for:
◦ SPARQL Protocol and RDF Query Language a semantic query language a query language for RDF(S) W3C Recommendation 15th January 2008
◦ provides a standard format for writing queries that target RDF data ◦ and a set of standard rules for processing queries and returning the results
http://www.w3.org/TR/rdf-sparql-query/
2
What is SPARQL
New version: 21st of March 2013: http://www.w3.org/TR/sparql11-overview/ Query language at: http://www.w3.org/TR/sparql11-query/ Implemented in various programming
languages
3
SPARQL now
4
RDF Statements
SubjectPredicate
author http://www.w3schools.com/RDF
Object
Jan Egil Refsnes
SPARQL searches for all sub-graphs that match the graph described by the triples in the query.
?subject ?predicate ?object
SELECT ?studentWHERE { ?student b:studies bmod:CS414 }
~ SQL – why?Which are the namespace prefixes?
5
A sample of SPARQL
In RDF:<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-
syntax-ns#" xmlns:cd="http://www.recshop.fake/cd#">
In SPARQL:PREFIX b: <http://...>
PREFIX bmod: <http://www2.warwick.ac.uk/fac/sci/dcs/teaching/material/>
The PREFIX keyword is SPARQL’s version of an xmlns:namespace declaration and works in basically the same way.
6
Prefixes & namespaces
Prefix IRI
rdf: http://www.w3.org/1999/02/22-rdf-syntax-ns#
rdfs: http://www.w3.org/2000/01/rdf-schema#
xsd: http://www.w3.org/2001/XMLSchema#
fn: http://www.w3.org/2005/xpath-functions#
sfn: http://www.w3.org/ns/sparql#
7
Frequently used namespaces
Prefixes are not optional in SPARQL!
Capitals of Europe: http://liris.cnrs.fr/~pchampin/spark/gmapv3.html
8
Dbpedia + Google maps mashup
http://dbpedia.org/sparql http://dbpedia.org/isparql/ http://dbpedia.org/snorql/ http://querybuilder.dbpedia.org/index.php
9
DbPedia Query points
http://dbpedia.org/resource/The_Lord_of_the_Rings http://dbpedia.org/resource/Berlin http://dbpedia.org/resource/Category:Cities_in_Englan
d http://dbpedia.org/resource/The_Beatles - Fluidops http://dbpedia.org/resource/Paul_McCartney http://dbpedia.org/resource/Category:English_musician
s http://dbpedia.org/resource/Semantic_Web http://dbpedia.org/resource/Tetris http://dbpedia.org/resource/SPARQL
10
Sample Resources
OpenLink Query point◦ http://demo.openlinksw.com/sparql
Semanticweb.org data:◦ http://data.semanticweb.org/snorql/
Diseases:◦ http://wifo5-03.informatik.uni-mannheim.de/diseasome/snorql/
Country statistics:◦ http://wifo5-03.informatik.uni-mannheim.de/factbook/snorql/
Nobelprize info:◦ http://data.nobelprize.org/snorql/
W3C’s list of endpoints:◦ http://www.w3.org/wiki/SparqlEndpoints
11
Other SPARQL query points
http://www.sparql.org/query.html http://sparql.org/sparql.html http://demo.openlinksw.com/sparql
◦ make sure to allow at ‘Sponging’ to ‘Retrieve remote RDF data for all missing source graphs’
http://librdf.org/query
12
Online SPARQL queries
Different syntax from XMLVariables: begin with ? or $ triples patterns enclosed within braces {}Result: variables after SELECT (~SQL)
13
SPARQL basics
PREFIX foo: PREFIX bar: SELECT … FROM … WHERE {…} GROUP BY … HAVING … ORDER BY … LIMIT … OFFSET … VALUES …
14
Anatomy of a Query
~ SQL
SELECT queries CONSTRUCT queries ASK queries DESCRIBE queries
15
4 Types of SPARQL Queries
PREFIX foaf: <http://xmlns.com/foaf/0.1/>SELECT ?x ?y ?zFROM <http://www.w3.org/People/Berners-Lee/card>WHERE { ?x ?y ?z }
16
PREFIX SELECT FROM WHERE
PREFIX b: <http://www2.warwick.ac.uk/rdf/>PREFIX bmod:
<http://www2.warwick.ac.uk/fac/sci/dcs/teaching/material/>
PREFIX foaf: <http://xmlns.com/foaf/0.1/>
SELECT ?name
WHERE { ?student b:studies bmod:CS328 .
?student foaf:name ?name }
17
Combining conditions
FOAF (Friend Of A Friend) experimental project using RDF, defining a
standardised vocabulary. Goal: make personal homepages machine-
readable & understandable, create an internet-wide connected database of people.
Defined at:http://xmlns.com/foaf/0.1/
18
FOAF
Idea: most personal homepages contain similar info.◦E.g, person’s name, living place, work place, details
on work of the moment, links to friends.◦define RDF predicates to represent them. ◦So: Pages: understood/ manipulated by computers.
So a db can be queried for: ◦“what projects are my friends working on?”, ◦“do any of my friends know the director of
BigCorp?” etc..
19
FOAF
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-
ns#" xmlns:foaf="http://xmlns.com/foaf/0.1#"> <foaf:Person> <foaf:name>Alexandra I. Cristea</foaf:name> <foaf:mbox rdf:resource="mailto:[email protected]"/> </foaf:Person> </rdf:RDF>
20
A sample FOAF RDF document
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#" xmlns:foaf="http://xmlns.com/foaf/0.1/"> <foaf:Person> <foaf:name>Peter Brusilovsky</foaf:name> <foaf:firstName>Peter</foaf:firstName> <foaf:surname>Brusilovsky</foaf:surname> <foaf:mbox_sha1sum>1fa1a097b3fae4594fa0ec18e2e3da11ac7e0349</foaf:mbox_sha1sum> <foaf:knows> <foaf:Person> <foaf:name>Alexandra I. Cristea</foaf:name> <foaf:mbox_sha1sum>dba246a0d9ec4a6238198f9a035a7a608ff227f8</foaf:mbox_sha1sum> <rdfs:seeAlso rdf:resource="http://www.dcs.warwick.ac.uk/~acristea/"/> </foaf:Person> </foaf:knows> </foaf:Person> </rdf:RDF>
21
FOAF: another example
PREFIX b: <http://www2.warwick.ac.uk/rdf/>PREFIX bmod: <
http://www2.warwick.ac.uk/fac/sci/dcs/teaching/material/>PREFIX foaf:http://xmlns.com/foaf/0.1/
SELECT ?module ?nameWHERE { ?student b:studies ?module . ?student foaf:name ?name }
22
Extracting multiple results
PREFIX b: <http://www2.warwick.ac.uk/rdf/>PREFIX bmod: <
http://www2.warwick.ac.uk/fac/sci/dcs/teaching/material/ >
PREFIX foaf:<http://xmlns.com/foaf/0.1/>
SELECT ?module ?name
WHERE { ?student b:studies ?module ; foaf:name ?name }
23
Using the same subject
SELECT ?module ?nameWHERE { ?student b:studies ?module . ?student b:studies bmod:CS328 ; foaf:name ?name }
is identical to:
SELECT ?module ?name
WHERE { ?student b:studies ?module , bmod:CS328 ;
foaf:name ?name }
24
Abbreviating multiple objects
SELECT ?student ?emailWHERE { ?student b:studies mod:CS328 . ?student foaf:mbox ?email }
PB: if a student does not have an e-mail address registered, with a foaf:mbox predicate, then the query will not match !!
25
Optional graph components
SELECT ?student ?emailWHERE { ?student b:studies mod:CS328 .OPTIONAL { ?student foaf:mbox ?email } }
OPTIONAL: match it if it can, but otherwise not reject the overall pattern.
26
Optional graph components
SPARQL basics SPARQL endpoints: dbpedia, foaf, others SPARQL type of queries:
◦ SELECT; CONSTRUCT; ASK; DESCRIBE SQL-like constructs:
◦ SELECT; FROM; WHERE; GROUP BY; HAVING; ORDER BY; LIMIT; OFFSET; VALUES
Some specific constructs:◦ PREFIX not optional; FROM not always used; OPTIONAL◦ We are going to look at more specific constructs
27
We have studied:
Try out at: http://www.sparql.org/query.html
PREFIX books: <http://example.org/book/>PREFIX dc: <http://purl.org/dc/elements/1.1/>SELECT $book WHERE {$book title $title } >> this gives an empty answer/ error.
Versus:PREFIX books: <http://example.org/book/>PREFIX dc: <http://purl.org/dc/elements/1.1/>SELECT $book WHERE {$book dc:title $title } 28
Prefixes are not optional
try out at : http://www.sparql.org/query.htmlWith or without optionalPREFIX books: <http://example.org/book/>PREFIX dc: <http://purl.org/dc/elements/1.1/>SELECT ?book ?author ?title ?dateWHERE { ?book dc:creator ?author . ?book dc:title ?title . ?book dc:date ?date }
PREFIX books: <http://example.org/book/>PREFIX dc: <http://purl.org/dc/elements/1.1/>SELECT ?book ?author ?title ?dateWHERE { ?book dc:creator ?author . ?book dc:title ?title . OPTIONAL { ?book
dc:date ?date } } 29
OPTIONAL query example
Try analysing RDF documents e.g.: http://athena.ics.forth.gr:9090/RDF/VRP/Examples/tap.rdf (on http://sparql.org/sparql.html or Protégé, or specific endpoints) Using SPARQL to find Concepts endpoints Dbpedia SNORQL query explorer (http://dbpedia.org/snorql/),
the BBC Backstage SPARQL Editor (http://bbc.openlinksw.com/sparql) or GeoSparql (http://geosparql.org/) to analyse content.
Listing DISTINCT RDF Types The following query lists the distinct rdf:types used in a dataset PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>SELECT DISTINCT ?typeWHERE { ?s rdf:type ?type } Listing OWL ClassesPREFIX owl: <http://www.w3.org/2002/07/owl#>SELECT ?classWHERE { ?class a owl:Class } 30
Analysing documents
SELECT ?module ?name ?phoneWHERE { ?student b:studies ?module . ?student foaf:name ?name .OPTIONAL { ?student b:contactpermission true .?student b:phone ?phone} }
SELECT ?module ?name ?ageWHERE { ?student b:studies ?module . ?student foaf:name ?name .OPTIONAL { ?student b:age ?age . FILTER (?age > 25) } }
31
More optional graph components …and filters:
SELECT ?student ?email ?homeWHERE { ?student b:studies mod:CS328 .OPTIONAL { ?student foaf:mbox ?email . ?student foaf:homepage ?home } }
SELECT ?student ?email ?homeWHERE { ?student b:studies mod:CS328 .OPTIONAL { ?student foaf:mbox ?email } .OPTIONAL { ?student foaf:homepage ?home } }
32
Further optional examples …
SELECT ?studentWHERE { ?student foaf:mbox ?email . { ?student b:studies mod:CS328 } UNION { ?student b:studies mod:CS909 } }
When patterns are combined using the UNION keyword, the resulting combined pattern will match if any of the subpatterns is matched.
33
Combining matches
Consider A and B as graph patterns. { A } UNION { B }
◦ Either / or graph patterns◦ Disjunction; include both the results of solving A and the
results of solving B { A } MINUS { B }
◦ “subtracted” graph patterns (SPARQL 1.1)◦ Negation; Solve A; Solve B; Include only those results from A
that are not compatible with any of the results from B.
34
Combining SPARQL Graph Patterns
All the previous queries: single RDF graphs. Often: multiple RDF graphs. RDF graphs: identified by IRI.
◦ Note: IRI that represents the graph does not have to be the actual IRI of the graph file
◦ although the program processing the query will need to somehow relate the IRI to an actual RDF graph stored somewhere.
35
Multiple graphs and the dataset
SELECT ?student ?email ?homeFROM <http://www2.warwick.ac.uk/rdf/student>WHERE { ?student b:studies mod:CS414 . OPTIONAL { ?student foaf:mbox ?email . ?student foaf:homepage ?home } }
By using several FROM declarations, you can combine several graphs in the dataset:
SELECT ?student ?email ?homeFROM <http://www2.warwick.ac.uk/rdf/student>FROM <http://www2.warwick.ac.uk/rdf/foaf>WHERE { ?student b:studies mod:CS414 . OPTIONAL { ?student foaf:mbox ?email . ?student foaf:homepage ?home } }
36
Stating the dataset
one (optional) default graph + any number of named graphs. FROM specifies default graph.
◦Many FROM keywords, the graphs are merged into the default graph.
Additionally: named graphs, w FROM NAMED.◦However, to match patterns you must use GRAPH
keyword to state which graph !
37
Multiple graphs
SELECT ?student ?email ?homeFROM NAMED <http://www2.warwick.ac.uk/rdf/student>FROM NAMED<http://www2.warwick.ac.uk/rdf/foaf>
WHERE { GRAPH <http://www2.warwick.ac.uk/rdf/student> { ?student b:studies mod:CS909 } . GRAPH <http://www2.warwick.ac.uk/rdf/foaf> { OPTIONAL { ?student foaf:mbox ?email . ?student foaf:homepage ?home } } }
38
Examples of named graphs
PREFIX brdf: <http://www2.warwick.ac.uk/rdf/>SELECT ?student ?email ?homeFROM NAMED <http://www2.warwick.ac.uk/rdf/student>FROM NAMED <http://www2.warwick.ac.uk/rdf/foaf>
WHERE { GRAPH brdf:student { ?student b:studies mod:CS909 } . GRAPH brdf:foaf{ OPTIONAL { ?student foaf:mbox ?email . ?student foaf:homepage ?home } } }
39
Abbreviation using prefixes
PREFIX brdf: <http://www2.warwick.ac.uk/rdf/>SELECT ?student ?email ?homeFROM <http://www2.warwickac.uk/rdf/student>FROM NAMED <http://www2.warwickac.uk/rdf/foaf>
WHERE { ?student b:studies mod:CS909 . GRAPH brdf:foaf{ OPTIONAL { ?studentfoaf:mbox?email . ?studentfoaf:homepage ?
home } } } 40
Using named and default graph together
the GRAPH can also be a variable.◦to query which graph in the dataset holds a
particular relationship, ◦or which graph to search based on data in
another graph.
not mandatory to declare all graphs◦even if specified, the dataset can be
overridden on a per-query basis.
41
Graph as a query
PREFIX brdf: <http://www2.warwick.ac.uk/rdf/>SELECT ?student ?graphWHERE { ?student b:studies mod:CS909 . GRAPH ?graph { ?student foaf:mbox ?email } }
output var graph holds graph URL (matching student to e-mail address).
Presumption: query processor has knowledge of a finite set of graphs + their locations
42
Which graph is it in?
PREFIX brdf: <http://www2.warwickac.uk/rdf/>SELECT ?student ?emailWHERE { ?student b:studies mod:CS909 . ?student rdfs:seeAlso ?graph . GRAPH ?graph { ?student foaf:mbox ?email } } Note: if student doesn’t have a rdfs:seeAlso property
which points to a graph holding their e-mail address, they will not appear in the result at all.
43
Re-using the graph reference
Specific SPARQL constructs: ◦ Revisited OPTIONAL, PREFIX, FROM - single◦ New ones: FILTER, UNION, MINUS, FROM –multiple,
FROM NAMED, GRAPH, GRAPH with prefix, GRAPH as (query-able + reusable) variable
More online/ offline queries:◦ Endpoints◦ Sparql.com◦ Protégé
Next: other SQL-like constructs; non-SQL-like behaviour: ASK, DESCRIBE, CONSTRUCT
44
We have learned:
SELECT ?name ?moduleWHERE { ?student b:studies ?module . ?student foaf:name ?name }ORDER BY ?name
SELECT ?name ?ageWHERE { ?student b:age ?age . ?student foaf:name ?name }ORDER BY DESC (?age) ASC (?name)
45
Sorting results of a query
SELECT ?name ?moduleWHERE { ?student b:studies ?module .?student foaf:name ?name }LIMIT 20
46
Limiting the number of results
SELECT ?name ?moduleWHERE { ?student b:studies ?module . ?student foaf:name ?name }ORDER BY ?nameOFFSET 10LIMIT 20
47
Extracting subsets of the results
SELECT ?studentWHERE { ?student b:studies ?module }
versus:
SELECT DISTINCT ?studentWHERE { ?student b:studies ?module }
48
Obtaining unique results
Is any student studying any module?ASK { ?student b:studies ?module }
Is any student studying CS909?ASK { ?student b:studies bmod:CS909 }
Is student 029389 studying CS909?ASK { bstu:029389 b:studies bmod:CS909 }
Is anyone whom 029389 knows, studying CS909?ASK {bstu:029389 foaf:knows ?x .
?x b:studies bmod: CS909 } Is any student aged over 30 studying CS909? ASK { ?student b:studies bmod:CS909 . ?student b:age ?age . FILTER { ?age > 30 } }
49
Obtaining a Boolean result
CONSTRUCT { ?student b:studyFriend ?friend }WHERE { ?student b:studies ?module . ?student foaf:knows ?friend . ?friend b:studies ?module } }
If there is more than one search result, the triples from each result are combined.
50
Constructing an RDF result
DESCRIBE ?country
Result is in RDF triples, in any RDF serialization
51
DESCRIBE queries
52
53
54
[1] Dean Allemang & Jim Hendler. 2008 Semantic Web for the working ontologist. Morgan Kaufmann publishers. ISBN 978-0-12-373556-0
[2] Eric Prud'hommeaux ; & Andy Seaborne . 2008 SPARQL Query Language for RDF , [Online] http://www.w3.org/TR/rdf-sparql-query/
55
References
Download Jena from http://jena.sourceforge.net/ Protégé 3.4 and above
http://protege.stanford.edu/ Fuseki:
http://jena.apache.org/documentation/serving_data/index.html
56
Tools to process
SPARQL language is used for querying RDF. SPARQL is not based on XML, but on SQL-
like syntax. Building blocks of SPARQL queries are graph
patterns that include variables. The result of the query will be the values the vars must take to match the RDF graph.
A SPARQL query can return results in several different ways, as determined by the query.
SPARQL queries can also be used for OWL querying.
57
Summary
58
Questions?