SPARQL and Linked Data

SPARQL - Query
Language for RDF

Fulvio Corno, Laura Farinetti
Politecnico di Torino
Dipartimento di Automatica e Informatica
e-Lite Research Group – http://elite.polito.it

The “new” Semantic Web vision
 To make data machine processable, we need:
 Unambiguous names for resources (that may also
bind data to real world objects): URIs
 A common data model to access, connect, describe
the resources: RDF
 Access to that data: SPARQL
 Define common vocabularies: RDFS, OWL, SKOS
 Reasoning logics: OWL, Rules

 “SPARQL will make a huge difference” (Tim
Berners-Lee, May 2006)
F. Corno, L. Farinetti - Politecnico di Torino 2

The Semantic Web timeline


SPARQL
 Queries are very important for distributed RDF
data
 Complex queries into the RDF data are often
necessary
 E.g.: “give me the (a,b) pair of resources, for which
there is an x such that (x parent a) and (b brother x)
holds” (i.e., return the uncles)
 This is the goal of SPARQL (Query Language
for RDF)


SPARQL
 SPARQL 1.0: W3C Recommendation January
15th, 2008
 SPARQL 1.1: W3C Working Draft January 5th,
2012
 SPARQL queries RDF graphs
 An RDF graph is a set of triples
 SPARQL can be used to express queries across
diverse data sources, whether the data is stored
natively as RDF or viewed as RDF via
middleware

SPARQL and RDF
 It is the triples that matter, not the serialization
 RDF/XML is the W3C recommendation but it not a
good choice because it allows multiple ways to
encode the same graph
 SPARQL uses the Turtle syntax, an
N-Triples extension


Turtle - Terse RDF Triple
Language N-Triples ⊂ Turtle ⊂ N3

 A serialization format for RDF
 A subset of Tim Berners-Lee and Dan
Connolly’s Notation 3 (N3) language
 Unlike full N3, doesn’t go beyond RDF’s graph model
 A superset of the minimal N-Triples format
 Turtle has no official status with any standards
organization, but has become popular amongst
Semantic Web developers as a human-friendly
alternative to RDF/XML

“Triple” or “Turtle” notation


<http://www.w3.org/People/EM/contact#me>
<http://www.w3.org/2000/10/swap/pim/contact#fullName>
"Eric Miller" .

<http://www.w3.org/2000/10/swap/pim/contact#mailbox>
<mailto:em@w3.org> .

<http://www.w3.org/2000/10/swap/pim/contact#personalTitle>
"Dr." .

<http://www.w3.org/1999/02/22-rdf-syntax-ns#type>
<http://www.w3.org/2000/10/swap/pim/contact#Person> .


(abbreviated)
w3people:EM#me contact:fullName "Eric Miller" .

w3people:EM#me contact:mailbox <mailto:em@w3.org> .

w3people:EM#me contact:personalTitle "Dr." .

w3people:EM#me rdf:type contact:Person .


Language
 Plain text syntax for RDF
 Based on Unicode
 Mechanisms for namespace abbreviation
 Allows grouping of triples according to
subject
 Shortcuts for collections


Language
 Simple triple:
subject predicate object .
:john rdf:label "John" .

 Grouping triples:
subject predicate object ; predicate object ...
:john
rdf:label "John" ;
rdf:type ex:Person ;
ex:homePage http://example.org/johnspage/ .


Prefixes
 Mechanism for namespace abbreviation
@prefix abbr: <URI>

 Example:
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>

 Default:
@prefix : <URI>

 Example:
@prefix : <http://example.org/myOntology#>


Identifiers
 URIs: <URI>
http://www.w3.org/1999/02/22-rdf-syntax-ns#

 Qnames (Qualified names)
 namespace-abbr?:localname
rdf:type
dc:title
:hasName
 Literals
 "string"(@lang)?(ˆˆtype)?

"John"
"Hello"@en-GB
"1.4"^^xsd:decimal

Blank nodes
 Simple blank node:
 [] or _:x :john ex:hasFather [] .
:john ex:hasFather _:x .

 Blank node as subject:
 [ predicate object ; predicate object ... ] .

[ ex:hasName "John"] .
[ ex:authorOf :lotr ;
ex:hasName "Tolkien"] .


Blank nodes


Collections
 ( object1 ... objectn )

:doc1 ex:hasAuthor (:john :mary) .

 Short for
:doc1 ex:hasAuthor
[ rdf:first :john;
rdf:rest [ rdf:first :mary;
rdf:rest rdf:nil ]
] .


Example
@prefix rdf: http://www.w3.org/1999/02/22-rdf-syntaxns# .
@prefix dc: <http://purl.org/dc/elements/1.1/> .
@prefix : <http://example.org/#> .

<http://www.w3.org/TR/rdf-syntax-grammar>
dc:title "RDF/XML Syntax Specification (Revised)" ;
:editor [
:fullName "Dave Beckett";
:homePage <http://purl.org/net/dajobe/>
] .


RDF Triplestores
 Basically, databases for triples

 Triplestores do not only store triples, but allow to
extract the “interesting” triples, via SPARQL
queries

Comparison
Relational database RDF Triplestore
 Data model  Data model
 Relational data (tables)  RDF graphs
 Data instances  Data instances
 Records in tables  RDF triples
 Query support  Query support
 SQL  SPARQL
 Indexing mechanisms  Indexing mechanisms
 Optimized for evaluating  Optimized for evaluating
queries as relational  Queries as graph patterns
expressions


SPARQL
 Uses SQL-like syntax
Prefix mechanism to abbreviate URIs

PREFIX dc: <http://purl.org/dc/elements/1.1/>
SELECT ?title
WHERE { <http://example.org/book/book1> dc:title
?title }

Variables to be returned
Query pattern (list of triple patterns)

FROM Name of the graph


SELECT
 Variables selection
?x
 Variables: ?string ?title
?name

 Syntax: SELECT var1,…,varn

SELECT ?name
SELECT ?x,?title


WHERE
 Graph patterns to match
 Set of triples
{ (subject predicate object .)* }
 Subject: URI, QName, Blank node, Literal,
Variable
 Predicate: URI, QName, Blank node, Variable
 Object: URI, QName, Blank node, Literal,
Variable


Graph patterns
 The pattern contains unbound symbols
 By binding the symbols (if possible), subgraphs
of the RDF graph are selected
 If there is such a selection, the query returns the
bound resources


Graph patterns

 E.g.: (subject,?p,?o)
 ?p and ?o are “unknowns”


Graph patterns SELECT ?p ?o
WHERE {subject ?p ?o}

 The triplets in WHERE define the graph pattern,
with ?p and ?o “unbound” symbols
 The query returns a list of matching p,o pairs


Example 1

SELECT ?cat, ?val
WHERE { ?x rdf:value ?val.
?x category ?cat }

 Returns:
[["Total Members",100],["Total Members",200],…,
["Full Members",10],…]

Example 2

SELECT ?cat, ?val
?x category ?cat.
FILTER(?val>=200). }

 Returns:
[["Total Members",200],…]


Example 3

SELECT ?cat, ?val, ?uri
?x category ?cat.
?al contains ?x.
?al linkTo ?uri }

 Returns:
[["Total Members",100,http://...)],…,]


Example 4

SELECT ?cat, ?val, ?uri
?x category ?cat.
OPTIONAL ?al contains ?x.
?al linkTo ?uri }

 Returns:
[["Total Members",100,http://...], …,
["Full Members",20, ],…,]

Other SPARQL Features
 Limit the number of returned results
 Remove duplicates, sort them,…
 Specify several data sources (via URI-s) within
the query (essentially, a merge)
 Construct a graph combining a separate pattern
and the query results
 Use datatypes and/or language tags when
matching a pattern


SPARQL use in practice
 Locally, i.e., bound to a programming
environments like Jena
 Jena is a Java framework for building Semantic Web
applications; provides an environment for RDF, RDFS
and OWL, SPARQL and includes a rule-based
inference engine
 Remotely, e.g., over the network or into a
database


Providing RDF on the Web
 RDF data usually resides in a RDF database
(triplestore)
 …how do we ‘put them out’ on the web?
 SPARQL endpoints (SPARQL query over HTTP)
 Direct connection to triplestore over HTTP


Example of SPARQL endpoint

Dataset

SPARQL query


Providing RDF
on the Web


Exposing RDF on the web
 Problem: usually HTML content and RDF data are
separate
Web
Page
(HTML)
RDF
data
(XML)


 Separate HTML content and RDF data
 Maintenance problem
 Both need to be managed separately
 RDF content and web content have much overlap
(redundancy)
 RDF/XML difficult to author: extra overhead
 Verification problem
 How to differences as content changes?
 Visibility problem
 Easy to ignore the RDF content (out of sight, out
of mind)

 Solution: embed RDF into web content using
RDFa
Web
Page
(HTML)
RDF
data
(XML)

‘Embed’ RDF into HTML


RDFa
 W3C Recommendation (October, 2008)
 Set of extensions to XHTML that allows to
annotate XHTML markup with semantics
 Uses attributes from XHTML meta and link
elements, and generalizes them so that they
are usable on all elements
A simple mapping is defined so that RDF
triples may be extracted


 RDFa: Resource Description Framework-in-attributes

 Extra (RDFa) markup is ignored by web browsers

XHTML + RDFa example
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML+RDFa 1.0//EN"
"http://www.w3.org/MarkUp/DTD/xhtml-rdfa-1.dtd">
<html xmlns="http://www.w3.org/1999/xhtml"
xmlns:foaf="http://xmlns.com/foaf/0.1/"
xmlns:dc="http://purl.org/dc/elements/1.1/"
version="XHTML+RDFa 1.0" xml:lang="en">
<head>
<title>John's Home Page</title>
<base href="http://example.org/john-d/" />
<meta property="dc:creator" content="Jonathan Doe" />
</head>
<body>
<h1>John's Home Page</h1>
<p>My name is <span property="foaf:nick">John D</span> and I like
<a href="http://www.neubauten.org/" rel="foaf:interest"
xml:lang="de">Einstürzende Neubauten</a>. </p>
<p> My <span rel="foaf:interest" resource="urn:ISBN:0752820907">
favorite book</span> is the inspiring
<span about="urn:ISBN:0752820907“><cite property="dc:title">
Weaving the Web</cite> by <span property="dc:creator">Tim
Berners-Lee</span></span> </p>
</body>
</html>

Automatic conversion to RDF/XML
<?xml version="1.0" encoding="UTF-8"?>
<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
xmlns:foaf="http://xmlns.com/foaf/0.1/"
xmlns:dc="http://purl.org/dc/elements/1.1/">
<rdf:Description rdf:about="http://example.org/john-d/">
<dc:creator xml:lang="en">Jonathan Doe</dc:creator>
<foaf:nick xml:lang="en">John D</foaf:nick>
<foaf:interest rdf:resource="http://www.neubauten.org/"/>
<foaf:interest>
<rdf:Description rdf:about="urn:ISBN:0752820907">
<dc:creator xml:lang="en">Tim Berners-Lee</dc:creator>
<dc:title xml:lang="en">Weaving the Web</dc:title>
</rdf:Description>
</foaf:interest>
</rdf:Description>
</rdf:RDF>


RDFa annotations
 Less than 5% of web pages have RDFa
annotations (Google, 2010)
 However, many organizations already publish
or consume RDFa
 Google, Yahoo
 Facebook, MySpace, LinkedIn
 Best Buy, Tesco, O’Reilly
 SlideShare, Digg
 WhiteHouse.gov, Library of Congress, UK
government
 Newsweek, BBC

RDFa is not the only solution …

Source: Peter Mika (Yahoo!), RDFa, 2011


Rich snippets
 Several solutions for embedding semantic data
in Web
 Three syntaxes known (by Google) as “rich
snippets”
 Microformats
 RDFa
 HTML microdata
 All three are supported by Google, while
microdata is the “recommended” syntax


First came microformats
 Microformats emerged around 2005
 Some key principles
 Startby solving simple, specific problems
 Design for humans first, machines second

 Wide deployment
 Used on billions of Web pages
 Formats exist for marking up atom feeds,
calendars, addresses and contact info, geo-
location, multimedia, news, products, recipes,
reviews, resumes, social relationships, etc.

Microformats example
<div class="vcard">
<a class="fn org url"
href="http://www.commerce.net/">CommerceNet</a>
<div class="adr">
<span class="type">Work</span>:
<div class="street-address">169 University Avenue</div>
<span class="locality">Palo Alto</span>,
<abbr class="region"
title="California">CA</abbr>  
<span class="postal-code">94301</span>
<div class="country-name">USA</div>
</div>
<div class="tel">
<span class="type">Work</span> +1-650-289-4040
</div>
<div>Email:
<span class="email">info@commerce.net</span>
</div>
</div>

Then came RDFa
 RDFa aims to bridge the gap between human
oriented HTML and machine-oriented RDF
documents
 Provides XHTML attributes to indicate machine
understandable information
 Uses the RDF data model, and Semantic Web
vocabularies directly


RDFa example

<div typeof="foaf:Person"
xmlns:foaf="http://xmlns.com/foaf/0.1/">
<p property="foaf:name">Alice Birpemswick</p>
<p>Email: <a rel="foaf:mbox"
href="mailto:alice@example.com">alice@example.com</a>
</p>
<p>Phone: <a rel="foaf:phone" href="tel:+1-617-555-7332">
+1 617.555.7332</a>
</p>
</div>


Last but not least, microdata
 Microdata syntax is based on nested groups of
name-value pairs
 HTML microdata specification includes
 An unambiguous parsing model
 An algorithm to convert microdata to RDF

 Compatible with the Semantic Web via
mappings


Microdata syntax
 Microdata properties
 Annotate an item with text-valued properties using the
“itemprop” attribute

<div itemscope>
<p>My name is <span itemprop="name">Daniel</span>.</p>
</div>


Microdata syntax
 Multiple values are ok
 As in RDF, you can have two properties, for the
same item (subject) with the same value (object)

<div itemscope>
<p>Flavors in my favorite ice cream:</p>
<ul>
<li itemprop="flavor">Lemon sorbet</li>
<li itemprop="flavor">Apricot sorbet</li>
</ul>
</div>


Microdata syntax
 Item types
 Correspond to classes in RDF

<section itemscope
itemtype="http://example.org/animals#cat">
<h1 itemprop="name">Hedral</h1>
<p itemprop="desc">Hedral is a male american
domestic shorthair, with a fluffy black fur with
white paws and belly.</p>
<img itemprop="img" src="hedral.jpeg" alt=""
title="Hedral, age 18 months">
</section>


Microdata syntax
 Global IDs
 Items may be given global identifiers, which are URLs
 They may be, but do not need to be Semantic Web URIs

<dl itemscope
itemtype="http://vocab.example.net/book"
itemid="urn:isbn:0-330-34032-8">
<dt>Title</dt>
<dd itemprop="title">The Reality Dysfunction</dd>
<dt>Author</dt>
<dd itemprop="author">Peter F. Hamilton</dd>
<dt>Publication date</dt>
<dd><time itemprop="pubdate" datetime="1996-01-26">
26 January 1996</time></dd>
</dl>


Schema.org
 Schema.org is one of a number of microdata
vocabularies
 It is a shared collection of microdata schemas for use
by webmasters
 Includes a type hierarchy, like an RDFS schema
 Starts with top-level Thing (which has four properties:
name, description, url, and image) and DataType types
 More specific types share properties with broader
types; for example, a Place is a more specific type of
Thing, and a TouristAttraction is a more specific type of
Place
 More specific items inherit the properties of their parent


Example
<div itemscope itemtype="http://schema.org/Movie">
<h1 itemprop="name"&g;Avatar</h1>
<div itemprop="director" itemscope
itemtype="http://schema.org/Person"> Director:
<span itemprop="name">James Cameron</span>(born
<span itemprop="birthDate">August 16, 1954)</span>
</div>
<span itemprop="genre">Science fiction</span>
<a href="../movies/avatar-theatrical-trailer.html"
itemprop="trailer">Trailer</a>
</div>


Current schema.org types


Schema.org full hierarchy

http://www.schema.org/docs/full.html


Item
examples


SPARQL use in practice
 Where to find meaningful RDF data to search?

 The Linked Data Project


The Linked Data Project
 A fundamental prerequisite of the Semantic Web is
the existence of large amounts of meaningfully
interlinked RDF data on the Web
 “To make the Semantic Web a reality, it is necessary
to have a large volume of data available on the Web
in a standard, reachable and manageable format. In
addition the relationships among data also need to be
made available. This collection of interrelated data on
the Web can also be referred to as Linked Data.
Linked Data lies at the heart of the Semantic Web:
large scale integration of, and reasoning on, data on
the Web.” (W3C)

 Linked Data is about using the Web to connect
related data that wasn’t previously linked, or
using the Web to lower the barriers to linking
data currently linked using other methods
 Linked Data is a set of principles that allows
publishing, querying and browsing of RDF data,
distributed across different servers


 Community effort to make various open datasets
available on the Web as RDF and to set RDF
links between data items from different datasets
 The datasets are published according to the
Linked Data principles and can therefore be
crawled by Semantic Web search engines and
navigated using Semantic Web browsers
 Supported by W3C
 Began early 2007
 http://linkeddata.org/home

The Web of Documents
 Analogy: a global filesystem
 Designed for human consumption
 Primary objects: documents
 Links between documents (or sub-parts)
 Degree of structure in objects: fairly low
 Semantics of content and links: implicit


The Web of Linked Data
 Analogy: a global database
 Designed for machines first, humans later
 Primary objects: things (or descriptions of things)
 Links between things
 Degree of structure in (descriptions of) things:
high
 Semantics of content and links: explicit


Linked Data example


Why publish Linked Data?
 Ease of discovery
 Ease of consumption
 Standards-based data sharing
 Reduced redundancy
 Added value
 Build ecosystems around your data/content


Linked Open Data cloud May 2007


DBpedia
 DBpedia is a community effort to extract
structured information from Wikipedia
and to make this information available on the
Web
 DBpedia allows to ask sophisticated queries
against Wikipedia, and to link other data sets
on the Web to Wikipedia data


GeoNames
 GeoNames is a geographical database that
contains over eight million geographical names
 Available for download free of charge under a
creative commons attribution license


Main contributors
 DBLP Computer science  Project Gutenberg Literary works
bibliography in the public domain
 Richard Cyganiak, Chris Bizer (FU  Piet Hensel, Hans Butschalowsky
Berlin) (FU Berlin)
 DBpedia Structured information  Revyu Community reviews about
from Wikipedia anything
 Universität Leipzig, FU Berlin,  Tom Heath, Enrico Motta (Open
OpenLink University)
 DBtune, Jamendo Creative  RDF Book Mashup Books from
Commons music repositories the Amazon API
 Yves Raimond (University of  Tobias Gauß, Chris Bizer (FU
London) Berlin)
 Geonames World-wide  US Census Data Statistical
geographical database information about the U.S.
 Bernard Vatant (Mondeca), Marc  Josh Tauberer (University of
Wick (Geonames) Pennsylvania), OpenLink
 Musicbrainz Music and artist  World Factbook Country
database statistics, compiled by CIA
 Frederick Giasson, Kingsley  Piet Hensel, Hans Butschalowsky
Idehen (Zitgist) (FU Berlin)


July 2007


August 2007


November 2007


February 2008


September 2008


March 2009


July 2009


September 2010


September 2011


Statistics on datasets
 http://ckan.net/group/lodcloud
 http://www4.wiwiss.fu-berlin.de/lodcloud/


Statistics on links between datasets


Linked Data success stories
 BBC Music
 Integrates information from MusicBrainz and
Wikipedia for artist/band infopages
 Information also available in RDF (in addition to web
pages)
 3rd party applications built on top of the BBC data
 BBC also contributes data back to the MusicBrainz
 Nytimes
 Maps its thesaurus of 1 million entity descriptions
(people, organisations, places, etc) to Dbpedia and
Freebase


Linked Data shopping list
 List of sites/datasets that the “community” would
like to see published as Linked Data
 This list may form the basis for some campaign/action
to encourage these data publishers to embrace
Linked Data

 http://linkeddata.org/linked-data-shopping-list


The Linked Data principles
(“expectations of behavior”)
 The Semantic Web isn't just about putting data
on the web. It is about making links, so that a
person or machine can explore the web of
data. With linked data, when you have some of
it, you can find other, related, data
 It is the unexpected re-use of information
which is the value added by the web
(Tim Berners-Lee)


The Linked Data principles
(“expectations of behavior”)
 Unambiguous identifiers for objects (resources)
 Use URIs as names for things
 Use the structure of the web
 Use HTTP URIs so that people can look up the
names
 Make is easy to discover information about an
object (resource)
 When someone lookups a URI, provide useful
information
 Link the object (resource) to related objects
 Include links to other URIs

Link to existing vocabularies
 For describing classes (categories) and properties
(relationships), try to re-use existing vocabularies
 Easier to interoperate if we’re talking the same
language!
 Many vocabularies/ontologies out there
 schema.org is a great place to start looking!
 Vocabs for products (Good Relations), people
(FOAF), social media (SIOC), places, events,
businesses, e-commerce, music, etc., you name it…
 If nothing relevant, you can create your own, but
make sure you…
 Publish it!
 Reconcile (map) it to other vocabularies, if you can

Link to other datasets
 Popular predicates for linking

 owl:equivalentClass  foaf:homepage
 owl:sameAs  foaf:topic
 rdfs:seeAlso  foaf:based_near
 skos:closeMatch  foaf:maker/foaf:made
 skos:exactMatch  foaf:page
 skos:related  foaf:primaryTopic

 Example:
http://dbpedia.org/resource/Canberra
owl:sameAs
http://rdf.freebase.com/rdf/en.canberra


Link to other Data Sets
(Wikicompany is a free content
licensed worldwide business
directory that anyone can edit)

(flickr wrappr extends DBpedia
with RDF links to photos posted
on flickr)


Linked Data – open issues
 Schema diversity & proliferation
 Quality of data is poor
 No kind of consistency is guarantees
 Issues with reliability of data end-points
 High down-time is not unusual
 There is no Service Level Agreement provided
 Querying of linked data is slow
 Data is distributed on the web
 Even single SPARQL endpoints can be slow
 Most end-points are experimental/research projects
with no resources for quality guarantees
 Licensing issues
 Majority of datasets carry no explicit open license

Linked Data tools
 Tools for Publishing Linked Data
 D2R Server: a tool for publishing relational databases
as Linked Data
 Triplify: transforms relational data into
RDF/LinkedData
 Pubby: a Linked Data frontend for SPARQL endpoints

 Tools for consuming Linked Data
 Semantic Web Browsers and Client Libraries
 Semantic Web Search Engines


Pubby
 Many triple stores and other SPARQL endpoints
can be accessed only by SPARQL client
applications that use the SPARQL protocol
 It cannot be accessed by the growing variety of
Linked Data clients
 Pubby is designed to provide a Linked Data
interface to those RDF data sources
 http://www4.wiwiss.fu-berlin.de/pubby/


Pubby


Linked Data browsers – Marbles
 http://marbles.sourceforge.net
 XHTML views of RDF data (SPARQL endpoint), caching,
predicate traversal


Linked Data browsers – RelFinder
 http://relfinder.dbpedia.org
 Explore & navigate relationships in a RDF graph


Linked Data browsers – gFacet
 http://semanticweb.org/wiki/GFacet
 Graph based visualisation & faceted filtering of RDF data


Linked Data browsers – Forest
 Front-end to FactForge and LinkedLifeData


FactForge and LinkedLifeData
 FactForge
 Integrates some of the most central LOD datasets
 General-purpose information (not specific to a domain)
 1.2B explicit plus 1B inferred statements
 The largest upper-level knowledge base
 http://www.FactForge.net/
 LinkedLifeData
 25 of the most popular life-science datasets
 2.7B explicit and 1.4B inferred triples
 http://www.LinkedLifeData.com


Linked Data browsers – Information Workbench
 http://iwb.fluidops.com/resource/Help:Start


http://it.ckan.net/


SPARQL query structure
 A SPARQL query includes, in order
 Prefix declarations, for abbreviating URIs
 A result clause, identifying what information to return
from the query
 The query pattern, specifying what to query for in the
underlying dataset
 Query modifiers: slicing, ordering, and otherwise
rearranging query results


SPARQL query structure
 A SPARQL query includes, in order
# prefix declarations
PREFIX foo: <http://example.com/resources/>
...
# result clause
SELECT ...
# query pattern
WHERE {
...
}
# query modifiers
ORDER BY ...


Dataset: Friend of a Friend (FOAF)
 FOAF is a standard RDF vocabulary for describing
people and relationships
 Tim Berners-Lee's FOAF information available at
http://www.w3.org/People/Berners-Lee/card

@prefix card: <http://www.w3.org/People/Berners-Lee/card#> .
@prefix foaf: <http://xmlns.com/foaf/0.1/> .
card:i foaf:name "Timothy Berners-Lee" .
<http://bblfish.net/people/henry/card#me>
foaf:name "Henry Story" .
<http://www.cambridgesemantics.com/people/about/lee>
foaf:name "Lee Feigenbaum" .
card:amy foaf:name "Amy van der Hiel" .
...


Example 1 – simple triple pattern
 In the graph http://www.w3.org/People/Berners-Lee/card,
find all subjects (?person) and objects (?name) linked
with the foaf:name predicate.
 Then return all the values of ?name.
 In other words, find all names mentioned in Tim Berners-
Lee’s FOAF file
PREFIX foaf:
<http://xmlns.com/foaf/0.1/>
SELECT ?name
WHERE {
?person foaf:name ?name .
}


SPARQL endpoints
 Accept queries and returns results via HTTP
 Generic endpoints queries any Web-accessible RDF data
 Specific endpoints are hardwired to query against
particular datasets
 The results of SPARQL queries can be returned in a
variety of formats:
 XML, JSON, RDF, HTML
 JSON (JavaScript Object Notation): lightweight computer
data interchange format; text-based, human-readable
format for representing simple data structures and
associative arrays


SPARQL endpoints
 This query is for an arbitrary bit of RDF data
(Tim Berners-Lee's FOAF file)
 => generic endpoint to run it
 Possible choices
 SPARQLer - General purpose processor - sparql.org
 http://sparql.org/sparql.html

 OpenLink's Virtuoso (Make sure to choose "Retrieve
remote RDF data for all missing source graphs")
 http://bbc.openlinksw.com/sparql/
 Redland’s Rasqal
 http://librdf.org/rasqal/


SPARQLer

SPARQL query

Dataset


OpenLink’s Virtuoso

Dataset

SPARQL query


Example 1 - simple triple pattern

PREFIX foaf:
SELECT ?name
WHERE {
}


Example 2 – multiple triple pattern
 Find all people in Tim Berners-Lee’s FOAF file that have
names and email addresses
 Return each person’s URI, name, and email address

 Multiple triple patterns retrieve multiple properties about
a particular resource
 SELECT * selects all variables mentioned in the query
PREFIX foaf:
SELECT *
WHERE {
?person foaf:mbox ?email .
}

Example 2 - multiple triple pattern


Example 3 – traversing a graph
 Find the homepage of anyone known by
Tim Berners-Lee


Example 3 – traversing a graph
PREFIX foaf: <http://xmlns.com/foaf/0.1/>
PREFIX card: <http://www.w3.org/People/Berners-Lee/card#>
SELECT ?homepage
FROM <http://www.w3.org/People/Berners-Lee/card>
WHERE {
card:i foaf:knows ?known .
?known foaf:homepage ?homepage .
}

 The FROM keyword specifies the target graph in the
query
 By using ?known as an object of one triple and the
subject of another, it is possible to traverse multiple links
in the graph

Dataset: DBPedia
 DBPedia is an RDF version of information from
Wikipedia
 Contains data derived from Wikipedia’s
infoboxes, category hierarchy, article abstracts,
and various external links
 Contains over 100 million triples
 Dataset: http://dbpedia.org/sparql/


Example 4 – exploring DBPedia
 Find 15 example concepts in the DBPedia
dataset
SELECT DISTINCT ?concept
WHERE {
?s a ?concept .
} LIMIT 15


Example 4 – exploring DBPedia
 LIMIT is a solution modifier that limits the
number of rows returned from a query
 SPARQL has two other solution modifiers
 ORDER BY for sorting query solutions on the value of
one or more variables
 OFFSET, used in conjunction with LIMIT and ORDER
BY to take a slice of a sorted solution set (e.g. for
paging)
 The SPARQL keyword a is a shortcut for the
common predicate rdf:type (class of a resource)
 The DISTINCT modifier eliminates duplicate
rows from the query results

Example 5 – basic SPARQL filters
 Find all landlocked countries with a population greater
than 15 million
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX type: <http://dbpedia.org/class/yago/>
PREFIX prop: <http://dbpedia.org/property/>
SELECT ?country_name ?population
WHERE {
?country a type:LandlockedCountries ;
rdfs:label ?country_name ;
prop:populationEstimate ?population .
FILTER (?population > 15000000) .
}

 FILTER constraints use boolean conditions to filter out
unwanted query results
 A semicolon (;) can be used to separate two triple
patterns that share the same subject

SPARQL filters
 Conditions on literal values
 Syntax
FILTER expression

 Examples
FILTER (?age > 30)
FILTER isIRI(?x)
FILTER !BOUND(?y)


SPARQL filters
 BOUND(var)
 true if var is bound in query answer
 false, otherwise
 !BOUND(var) enables negation-as-failure
 Testing types
 isIRI(A): A is an “Internationalized Resource
Identifier”
 isBLANK(A): A is a blank node
 isLITERAL(A): A is a literal


SPARQL filters
 Comparison between A = B
A != B
RDF terms
A = B
A != B
 Comparison between A <= B
A >= B
Numeric and Date types A < B
A > B

 Boolean AND/OR A && B
A || B

A + B
 Basic arithmetic A - B
A * B
A / B

Example 5 – basic SPARQL filters

 Note all the translated
duplicates in the results
 How can we deal with
that?


Example 6 – SPARQL filters
 Find me all landlocked countries with a
population greater than 15 million (revisited),
with the highest population country first
PREFIX type: <http://dbpedia.org/class/yago/>
PREFIX prop: <http://dbpedia.org/property/>
SELECT ?country_name ?population
WHERE {
?country a type:LandlockedCountries ;
rdfs:label ?country_name ;
prop:populationEstimate ?population .
FILTER (?population > 15000000 &&
langMatches(lang(?country_name), "EN")) .
} ORDER BY DESC(?population)


Example 6 – SPARQL filters
 lang extracts a literal’s language tag, if any
 langMatches matches a language tag against a
language range


Dataset: Jamendo
 Jamendo is a community collection of music all
freely licensed under Creative Commons
licenses
 http://www.jamendo.com/it/

 DBTune.org hosts a queryable RDF version of
information about Jamendo's music collection
 Data on thousands of artists, tens of thousands of
albums, and nearly 100,000 tracks
 http://dbtune.org/jamendo/store/


Example 7 – the wrong way
 Find all Jamendo artists along with their image,
home page, and the location they’re near

PREFIX mo: <http://purl.org/ontology/mo/>
SELECT ?name ?img ?hp ?loc
WHERE {
?a a mo:MusicArtist ;
foaf:name ?name ;
foaf:img ?img ;
foaf:homepage ?hp ;
foaf:based_near ?loc .
}


Example 7 – DBTune SPARQL
endpoint

http://dbtune.org/jamendo/store/

 Jamendo has information on about 3,500 artists
 Trying the query we only get 2,667 results. What's
wrong?

Example 7 – the right way
 Not every artist has an image, homepage, or location!
 OPTIONAL tries to match a graph pattern, but doesn't
fail the whole query if the optional match fails
 If an OPTIONAL pattern fails to match for a particular
solution, any variables in that pattern remain unbound
(no value) for that solution
PREFIX mo: <http://purl.org/ontology/mo/>
SELECT ?name ?img ?hp ?loc
WHERE {
?a a mo:MusicArtist ;
foaf:name ?name .
OPTIONAL { ?a foaf:img ?img }
OPTIONAL { ?a foaf:homepage ?hp }
OPTIONAL { ?a foaf:based_near ?loc }
}


Dataset: GovTrack
 GovTrack provides SPARQL access to data on
the U.S. Congress
 Contains over 13,000,000 triples about
legislators, bills, and votes
 http://www.govtrack.us/


Example 8 – querying alternatives
 Find Senate bills that either John McCain or Barack
Obama sponsored and the other cosponsored
PREFIX bill: <http://www.rdfabout.com/rdf/schema/usbill/>
PREFIX dc: <http://purl.org/dc/elements/1.1/>
SELECT ?title ?sponsor ?status
WHERE {
{ ?bill bill:sponsor ?mccain ; bill:cosponsor ?obama . }
UNION
{ ?bill bill:sponsor ?obama ; bill:cosponsor ?mccain . }
?bill a bill:SenateBill ;
bill:status ?status ;
bill:sponsor ?sponsor ;
dc:title ?title .
?obama foaf:name "Barack Obama" .
?mccain foaf:name "John McCain" .
}


Example 8 – GovTrack specific
endpoint

http://www.govtrack.us/developers/rdf.xpd

 The UNION keyword forms a disjunction of two graph
patterns: solutions to both sides of the UNION are
included in the results

RDF datasets
 All queries so far have been against a single graph
 In SPARQL this is known as the default graph
 RDF datasets are composed of a single default graph
and zero or more named graphs, identified by a URI
 Named graphs can be specified with one or more
FROM NAMED clauses, or they can be hardwired into a
particular SPARQL endpoint
 The SPARQL GRAPH keyword allows portions of a
query to match against the named graphs in the RDF
dataset
 Anything outside a GRAPH clause matches against the
default graph

Dataset: semanticweb.org
 data.semanticweb.org hosts RDF data regarding
workshops, schedules, and presenters for the
International Semantic Web (ISWC) and European
Semantic Web Conference (ESWC) series of events
 Presents data via FOAF, SWRC, and iCal ontologies
 The data for each individual ISWC or ESWC event is
stored in its own named graph
 i.e., there is one named graph per conference event contained in
this dataset
 http://data.semanticweb.org/


Example 9 – querying named graphs

 Find people who have been involved with at
least three ISWC or ESWC conference events
SELECT DISTINCT ?person
WHERE {
GRAPH ?g1 { ?person a foaf:Person }
FILTER(?g1 != ?g2 && ?g1 != ?g3 && ?g2 != ?g3) .
}


Example 9 – querying named graphs

 The GRAPH ?g construct allows a pattern to match
against one of the named graphs in the RDF dataset
 The URI of the matching graph is bound to ?g
(or whatever variable was actually used)
 The FILTER assures that we’re finding a person who
occurs in three distinct graphs
 The Web interface used for this SPARQL query defines
the foaf: prefix, which is why it is omitted here


Data.semanticweb.org specific
SPARQL endpoint
http://data.semanticweb.org/snorql/


SPARQL 1.1 extensions
 Projected expressions
 Adds the ability for query results to contain values
derived from constants, function calls, or other
expressions in the SELECT list
 Aggregates
 Adds the ability to group results and calculate
aggregate values (e.g. count, min, max, avg, sum, …)
 Sub-queries
 Allows one query to the nested within another


SPARQL 1.1 extensions
 Negation
 Includes improved language syntax for querying
negations
 Property paths
 Adds the ability to query arbitrarily lenght path
through a graph via a regular-expression-like syntax
known as property paths
 Basic federated query
 Defines a mechanism for splitting a single query
among multipleSPAQL endpoints and combining
together the results from each

Exercises - RDF
@prefix : <http://example.org/data#> .
@prefix ont: <http://example.org/myOntology#> .
@prefix vcard: <http://www.w3.org/2001/vcard-rdf/3.0#> .

:john
vcard:FN "John Smith" ;
vcard:N [
vcard:Given "John" ;
vcard:Family "Smith" ] ;
ont:hasAge 32 ;
ont:marriedTo :mary .
:mary
vcard:FN "Mary Smith" ;
vcard:N [
vcard:Given "Mary" ;
vcard:Family "Smith" ] ;
ont:hasAge 29 .


SPARQL query – exercise 1
 Return the full names of all people in the graph
PREFIX vCard:
<http://www.w3.org/2001/vcardrdf/3.0#>
SELECT ?fullName
WHERE {?x vCard:FN ?fullName}

 Result
fullName
===============
“John Smith”
“Mary Smith”


 Return the relation between John and Mary
PREFIX : <http://example.org/data#>
SELECT ?p
WHERE { :john ?p :mary }

 Result
p
=================
<http://example.org/myOntology#marriedTo>


 Return the spouse of a person whose name is
John Smith
PREFIX vCard:
<http://www.w3.org/2001/vcard-rdf/3.0#>
PREFIX ont: <http://example.org/myOntology#>
SELECT ?y
WHERE {?x vCard:FN "John Smith".
?x ont:marriedTo ?y}
 Result
y
=================
<http://example.org/data#mary>


 Return the name and the first name of all people
in the knowledge base
PREFIX vCard:
<http://www.w3.org/2001/vcard-rdf/3.0#>
SELECT ?name, ?firstName
WHERE {?x vCard:N ?name .
?name vCard:Given ?firstName}

 Result
name firstName
========================
“John Smith” "John"
“Mary Smith” "Mary"


 Return all people over 30 in the knowledge base

PREFIX ont: <http://example.org/myOntology#>
SELECT ?x
WHERE {?x ont:hasAge ?age .
FILTER(?age > 30)}

 Result
x
=================
<http://example.org/data#john>


FROM

 Select RDF graph (= dataset) to be queried
 In case of multiple FROM clauses, graphs are
merged
 Example

SELECT ?name
FROM <http://example.org/foaf/aliceFoaf>
WHERE { ?x foaf:name ?name }


 Graph http://example.org/bob

_:a foaf:name "Bob" .
_:a foaf:mbox <mailto:bob@oldcorp.example.org> .

 Graph http://example.org/alice
_:a foaf:name "Alice" .
_:a foaf:mbox <mailto:alice@work.example> .


 Return the names of people in both graphs
SELECT ?src ?name
FROM NAMED <http://example.org/alice>
FROM NAMED <http://example.org/bob>
WHERE
{ GRAPH ?src { ?x foaf:name ?name } }

 Result src name
=======================================
<http://example.org/bob> "Bob"
<http://example.org/alice> "Alice"


References
 W3C, “Introduction to the Semantic Web”
 http://www.w3.org/2006/Talks/0524-Edinburgh-IH/
 Lee Feigenbaum, “SPARQL By Example”
 http://www.cambridgesemantics.com/2008/09/sparql-by-example
 Valentina Tamma, “Chapter 4: SPARQL”
 http://www.csc.liv.ac.uk/~valli/Comp318/PDF/SPARQL.pdf
 Tom Heath, “An Introduction to Linked Data”
 http://tomheath.com/slides/2009-02-austin-linkeddata-tutorial.pdf
 HTML Microdata
 http://www.w3.org/TR/microdata
 Schema.org
 http://schema.org


License
This work is licensed under the Creative
Commons Attribution-Noncommercial-
Share Alike 3.0 Unported License.
To view a copy of this license, visit
http://creativecommons.org/licenses/by-
nc-sa/3.0/ or send a letter to Creative
Commons, 171 Second Street, Suite 300,
San Francisco, California, 94105, USA.


SPARQL and Linked Data

More Related Content

What's hot (19)

Similar to SPARQL and Linked Data (20)

Recently uploaded (20)

SPARQL and Linked Data