Linked Data Tutorial
@ Vlaams Theater Instituut
Brussels, 03/2009
bernhard.haslhofer@univie.ac.at
Bernhard Haslhofer, Linked Data Tutorial
Friday, March 6, 2009
Contents
• Motivation for Linked Open Data (LOD)
• Base Technologies
• The Linked Data Principles
• Publishing Solutions and Tools
• Vocabularies & Interoperability Issues
• Demos + Discussion
Bernhard Haslhofer, Linked Data Tutorial 2
Friday, March 6, 2009
How do we look up information about things?
Bernhard Haslhofer, Linked Data Tutorial 3
Friday, March 6, 2009
... about this book?
Bernhard Haslhofer, Linked Data Tutorial 4
Friday, March 6, 2009
... about this play?
Bernhard Haslhofer, Linked Data Tutorial 5
Friday, March 6, 2009
... about these concepts?
Bernhard Haslhofer, Linked Data Tutorial 6
Friday, March 6, 2009
... about things?
...that book? ...this play? ...these concepts?
we look up their names
Metamorphosis Effi Briest Love / Hate
and see what we can find out
Bernhard Haslhofer, Linked Data Tutorial 7
Friday, March 6, 2009
If information is not available on the Web, people
tend to ignore it.
Bernhard Haslhofer, Linked Data Tutorial 14
Friday, March 6, 2009
Current Situation
Web
Web Server
JDBC HTML
Database
Bernhard Haslhofer, Linked Data Tutorial 15
Friday, March 6, 2009
The current Web is a Web of Documents intended
for human interpretation.
Bernhard Haslhofer, Linked Data Tutorial 16
Friday, March 6, 2009
Database
Database
Database
The data are still locked in closed silos.
Database
Database Database
Bernhard Haslhofer, Linked Data Tutorial 17
Friday, March 6, 2009
Other applications can not access and process
these data unless...
Bernhard Haslhofer, Linked Data Tutorial 18
Friday, March 6, 2009
Existing Approaches
CD-ROM FTP
REST
OAI-PMH
Web Services
SRU/SRQ SOAP
OAI-ORE
RSS WSDL
Atom UDDI
JSON CD-ROM
RPC
RMI CORBA
Z39.50 DCOM
Bernhard Haslhofer, Linked Data Tutorial 19
Friday, March 6, 2009
The linked data vision is to...
Bernhard Haslhofer, Linked Data Tutorial 20
Friday, March 6, 2009
Documents
Data
Database
Bernhard Haslhofer, Linked Data Tutorial
Friday, March 6, 2009
• open the data silos and get rid of repository-centric
mindset
• publish data of public interest on the Web
• in a way that other applications can access and
interpret the data
• using common Web technologies
Bernhard Haslhofer, Linked Data Tutorial 22
Friday, March 6, 2009
make it possible for applications to look up
(meta)data ...
Bernhard Haslhofer, Linked Data Tutorial 23
Friday, March 6, 2009
... about things
...that book? ...this play? ...these concepts?
by looking up their names
http://dbpedia.org/resource/
Love
http://dbpedia.org/resource/ http://dbpedia.org/resource/
The_Metamorphosis Effi_Briest
http://dbpedia.org/resource/
Hatred
on the Web
Bernhard Haslhofer, Linked Data Tutorial 24
Friday, March 6, 2009
To do so, we need some technology
Bernhard Haslhofer, Linked Data Tutorial 25
Friday, March 6, 2009
Uniform Resource Identifier (URI)
• Names (Identifiers) for resources in an open
environment
• dereferencable URI = HTTP URI (URL)
http://dbpedia.org/resource/
Hallstatt_culture
Bernhard Haslhofer, Linked Data Tutorial 26
Friday, March 6, 2009
Resource Description Framework (RDF)
• a model for representing metadata on the Web
• in the form of statements (triples)
“The Hallstatt culture
t
abstrac
: was ...”
p
dbppro
rdfs: label
http://dbpedia.org/resource/
“Hallstatt culture”
Hallstatt_culture
skos:s
ubject
http://
dbpedia.org/resource/
Category:Iron_Age_Europe
Bernhard Haslhofer, Linked Data Tutorial 27
Friday, March 6, 2009
RDF/XML, N3, Turtle, etc.
• for exchanging RDF data
• serialization & de-serialization
<rdf:Description rdf:about=” http://dbpedia.org/resource/Hallstatt_culture”>
<dbpprop:abstract>The Hallstatt culture was...</dbpprop:abstract>
<skos:subject rdf:resource=”http://dbpedia.org/resource/
Category:Iron_Age_Europe”>
<rdfs:label>Hallstatt culture</rdfs:label>
</rdf:Description>
Bernhard Haslhofer, Linked Data Tutorial 28
Friday, March 6, 2009
RDFS & OWL
• languages for describing vocabularies
rdf: Property
rdf: type rdf: type
rdf: type
dbpprop: abstract skos: subject rdfs: label
Bernhard Haslhofer, Linked Data Tutorial 29
Friday, March 6, 2009
Simple Knowledge Organization System (SKOS)
• a language for describing controlled vocabularies
“Iron Age
“Hallstatt Period” “Hallstatt Culture”
Europe”
skos:prefLabel skos:prefLabel skos:altLabel
“La Tène Period”
skos:narrower
dbpedia:Category:
xyz: HallstattCulture
skos:prefLabel
Iron_Age_Europe
skos:narrower xyz: LateneCulture
Bernhard Haslhofer, Linked Data Tutorial 30
Friday, March 6, 2009
SPARQL
• a query language & protocol for accessing RDF data
via the Web
SELECT ?uri
WHERE {
?uri skos:subject <http://dbpedia.org/resource/Category:Iron_Age_Europe>.
}
Bernhard Haslhofer, Linked Data Tutorial 31
Friday, March 6, 2009
The vision is becoming reality...
Bernhard Haslhofer, Linked Data Tutorial 32
Friday, March 6, 2009
http://en.wikipedia.org/wiki/Hallstatt_culture
<http://dbpedia.org/resource/Hallstatt_culture >
p:abstract “Die Hallstattzeit bezeichnet...” @de
p:abstract “The Hallstatt culture was...” @en
rdfs:label “Hallstatt culture”@en;
rdfs:label “Hallstattzeit” @de;
rdfs:label “Civilisation de Hallstatt” @fr;
rdfs:label “ ” @jp;
skos:subject dbpedia:Category:Iron_Age_Europe;
...
Bernhard Haslhofer, Linked Data Tutorial 33
Friday, March 6, 2009
<http://dbpedia.org/resource/Hallstatt_culture >
...
p:hasPhotoCollection <http://www4.wiwiss.fu-berlin.de/flickrwrappr/
photos/Hallstatt_culture>;
...
Bernhard Haslhofer, Linked Data Tutorial 34
Friday, March 6, 2009
Library of Congress Subject Headings
Bernhard Haslhofer, Linked Data Tutorial 35
Friday, March 6, 2009
The LOD idea in brief
• expose data on the Web, not just documents
• interlink these data with those of other data sources
Bernhard Haslhofer, Linked Data Tutorial 37
Friday, March 6, 2009
LOD benefits
• other humans and applications can
• easily access your data using Web technologies
• follow the links in order to obtain further contextual
information
• links to your data and search engine indices can
increase the visibility of your data
Bernhard Haslhofer, Linked Data Tutorial 38
Friday, March 6, 2009
“Stop hugging
your data”
Sir Tim Berners-Lee, 2009
(c) Paul Miller, http://www.slideshare.net/
cloudofdata/toward-the-data-cloud
Friday, March 6, 2009
Contents
• Motivation for Linked Open Data (LOD)
• Base Technologies
• The Linked Data Principles
• Publishing Solutions and Tools
• Vocabularies & Interoperability Issues
• Demos + Discussion
Bernhard Haslhofer, Linked Data Tutorial 40
Friday, March 6, 2009
The Semantic Web Layer Cake
User Interface & Applications
Trust
Proof
Unifying Logic
Vocabulary: Ontology: Rules:
SKOS OWL RIF
Query:
SPARQL
Crypto
RDF-S
Data Interchange: RDF
XML
URI Unicode
Bernhard Haslhofer, Linked Data Tutorial 41
Friday, March 6, 2009
From a Database point of view
Database Domain Semantic Web
Relational Model
URI, RDF
(Tables)
SQL DDL
RDFS, OWL
(Create Table table-name ...)
SQL Query Language
SPARQL
(Select * from ...)
Bernhard Haslhofer, Linked Data Tutorial 42
Friday, March 6, 2009
Uniform Resource Identifiers (URI)
• Unambiguous name for “something”
• for a digital resource
http://dbpedia.org/resource/
• for a concept within a vocabulary Hallstatt_culture
• etc..
• “A URI is a compact sequence of characters that
identifies an abstract or physical resource”
[RFC 3986]
http://dbpedia.org/resource/
Hallstatt
Bernhard Haslhofer, Linked Data Tutorial 43
Friday, March 6, 2009
Uniform Resource Identifiers (URI)
• “Uniform”:
• different types of resource identifiers in the same context
• common syntactic conventions
• “Resource”:
• whatever might be identified by a URI
• “Identifier”:
• distinguish one resource from other ones
Bernhard Haslhofer, Linked Data Tutorial 44
Friday, March 6, 2009
Uniform Resource Identifiers (URI)
• An URI can be classified as locator (URL), name
(URN), or both
• Uniform Resource Locator (URL)
• a means for locating the resource by describing its
primary access mechanism (e.g., http://example.com)
• Uniform Resource Name (URN)
• a means for naming the resource (e.g.,
urn:example.com:animal)
Bernhard Haslhofer, Linked Data Tutorial 45
Friday, March 6, 2009
Uniform Resource Identifiers (URI)
• Generic syntax - a hierarchical sequence of
components
URI = scheme “:” hier-path [ “?” query ] [ “#” fragment ]
foo://example.com:8042/over/there?name=ferret#nose
_/ ________________/_________/ _________/ __/
| | | | |
scheme authority path query fragment
| ______________________|_
//
urn:example:animal:ferret:nose
Bernhard Haslhofer, Linked Data Tutorial 46
Friday, March 6, 2009
Uniform Resource Identifiers (URI)
• Example URIs:
ftp://ftp.is.co.za/rfc/rfc1808.txt
http://www.ietf.org/rfc/rfc2396.txt
ldap://[2001:db8::7]/c=GB?objectClass?one
mailto:John.Doe@example.com
news:comp.infosystems.www.servers.unix
tel:+1-816-555-1212
urn:oasis:names:specification:docbook:dtd:xml:4.1.2
Bernhard Haslhofer, Linked Data Tutorial 47
Friday, March 6, 2009
Internationalized Resource Identifiers (IRI)
• can we use non-ASCII characters in URIs (e.g.,
Umlaut, Chinese, Arabic, Hebrew, etc...)?
• extend the syntax of URI so that a much wider
repertoire of characters can be used
• uses Universal Character Set (ISO 10646)
• mapping from IRIs to URIs
• IRIs can be used wherever URIs are allowed
Bernhard Haslhofer, Linked Data Tutorial 48
Friday, March 6, 2009
The Semantic Web Layer Cake
User Interface & Applications
Trust
Proof
Unifying Logic
Vocabulary: Ontology: Rules:
SKOS OWL RIF
Query:
SPARQL
Crypto
RDF-S
Data Interchange: RDF
XML
URI Unicode
Bernhard Haslhofer, Linked Data Tutorial 49
Friday, March 6, 2009
Resource Description Framework (RDF)
• The basic structural element of RDF is the statement
rdfs: label
http://dbpedia.org/resource/
“Hallstatt culture”
Hallstatt_culture
skos:s
ubject
http://dbpedia.org/resource/
Category:Iron_Age_Europe
Subject Predicate Object
Subject Property Value
Bernhard Haslhofer, Linked Data Tutorial 50
Friday, March 6, 2009
Resource Description Framework (RDF)
• The subject of a statement is always a (URI) resource
• The predicate of a statement is always a (URI)
resource
• The object of a statement can be a resource (URI) or
a typed literal
• An RDF statement forms a triple
• Triples can be merged into a set of triples, forming a
directed labeled graph
Bernhard Haslhofer, Linked Data Tutorial 51
Friday, March 6, 2009
Resource Description Framework (RDF)
• Literals can be plain or typed using arbitrary datatypes
• It is recommended to use XML Schema datatypes
[XMLS2]
rdfs: label
http://dbpedia.org/resource/
“Hallstatt”
Hallstatt
geo:lo
ng
“13.646667”^^<http://
www.w3.org/2001/
XMLSchema#float>
Bernhard Haslhofer, Linked Data Tutorial 52
Friday, March 6, 2009
Resource Description Framework (RDF)
• Support for multilingual labels
• Plain literals may have a language tag, as defined by
RFC 3066
“ (
l
: labe )”@jp
rdfs
rdfs: label
http://dbpedia.org/resource/
“Hallstatt”@de
Hallstatt
rdfs:la
bel
“Гальштат”@ru
Bernhard Haslhofer, Linked Data Tutorial 53
Friday, March 6, 2009
Resource Description Framework (RDF)
• Blank nodes can be used to model structured
information that needs no URI by itself
• they must still be distinguishable within an RDF graph.
This is done using blank node identifiers
http://www.ex.com/
http://www.ex.com/#address
staff#85740
http://www.ex.com/#zipcode
http://www.ex.com/#city
http://www.ex.com/#street http://www.ex.com/#state
quot;01730quot;
quot;Bedfordquot;
quot;1501 Grant
quot;Massachusettsquot;
Avenuequot;
Bernhard Haslhofer, Linked Data Tutorial 54
Friday, March 6, 2009
Resource Description Framework (RDF)
• Containers allow to describe groups of things
• Bag (unordered set), Seq (ordered set), Alt (choice)
http://www.ex.com/ http://www.w3.org/1999/02/22-rdf-
courses#6.001 syntax-ns#Bag
http://www.ex.com/students#students http://www.ex.com/students#Amy
http://www.w3.org/1999/02/22-rdf-syntax-ns#type
http://www.w3.org/1999/02/22-rdf-syntax-ns#_1
http://www.ex.com/
http://www.w3.org/1999/02/22-rdf-syntax-ns#_2 students#Mohamed
http://www.w3.org/1999/02/22-rdf-syntax-ns#_3
http://www.w3.org/1999/02/22-rdf-syntax-ns#_4 http://www.ex.com/students#Johann
blank node
http://www.ex.com/students#Maria
Bernhard Haslhofer, Linked Data Tutorial 55
Friday, March 6, 2009
Resource Description Framework (RDF)
• Collections allow the definition of closed containers
• define the exact set of items in a collection
http://www.ex.com/
courses#6.001
http://www.ex.com/students#students
http://www.w3.org/1999/02/22-rdf-syntax-ns#first
http://www.ex.com/students#Amy
http://www.w3.org/1999/02/22-rdf-syntax-ns#rest
http://www.w3.org/1999/02/22-rdf-syntax-ns#first
http://www.ex.com/
students#Mohamed
http://www.w3.org/1999/02/22-rdf-syntax-ns#rest
http://www.w3.org/1999/02/22-rdf-syntax-ns#first
http://www.ex.com/students#Johann
http://www.w3.org/1999/02/22-rdf-syntax-ns#rest
http://www.w3.org/1999/02/22-rdf-
syntax-ns#nil
Bernhard Haslhofer, Linked Data Tutorial 56
Friday, March 6, 2009
Resource Description Framework (RDF)
• RDF can be serialized using various syntax formats:
<?xml version=”1.0”?>
<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:dbpprop=”http://dbpedia.org/property/” .... >
<rdf:Description rdf:about=” http://dbpedia.org/resource/Hallstatt_culture”>
<dbpprop:abstract>The Hallstatt culture was...</dbpprop:abstract>
<skos:subject rdf:resource=” http://dbpedia.org/resource/
Category:Iron_Age_Europe”>
<rdfs:label>Hallstatt culture</rdfs:label>
RDF/XML
</rdf:Description>
Bernhard Haslhofer, Linked Data Tutorial 57
Friday, March 6, 2009
Resource Description Framework (RDF)
• RDF can be serialized using various syntax formats:
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>
@prefix rdfs: <http://dbpedia.org/about/html/http://www.w3.org/2000/01/rdf-schema#>
@prefix dbpprop: <http://dbpedia.org/property/>
<http://dbpedia.org/resource/Hallstatt_culture>
dbpprop:abstract “The Hallstatt culture was...” ;
skos:subject < http://dbpedia.org/resource/Category:Iron_Age_Europe > ;
Turtle / N3
rdfs:label “Hallstatt culture” .
Bernhard Haslhofer, Linked Data Tutorial 58
Friday, March 6, 2009
Resource Description Framework (RDF)
• Reification
• allows to describe other statements; i.e. make statements
about other statements
• rarely used in practice
Bernhard Haslhofer, Linked Data Tutorial 59
Friday, March 6, 2009
The Semantic Web Layer Cake
User Interface & Applications
Trust
Proof
Unifying Logic
Vocabulary: Ontology: Rules:
SKOS OWL RIF
Query:
SPARQL
Crypto
RDF-S
Data Interchange: RDF
XML
URI Unicode
Bernhard Haslhofer, Linked Data Tutorial 60
Friday, March 6, 2009
RDF Schema (RDFS)
• extends RDF with the possibility to define classes and
associated properties
• allows RDF applications to agree on a common data
description vocabulary
• is implemented on top of RDF - each valid RDFS
document is a valid RDF document
Bernhard Haslhofer, Linked Data Tutorial 61
Friday, March 6, 2009
RDF Schema (RDFS)
• classes and subclass relationships
rdfs: Class
rdf:type
rdf:type
yago:
yago: City CitiesAndTownsInUpperAustria
rdfs: subClassOf
rdf:type
http://dbpedia.org/resource/
Hallstatt
Bernhard Haslhofer, Linked Data Tutorial 62
Friday, March 6, 2009
RDF Schema (RDFS)
• properties and subPropertyOf relationships
rdf: Property
rdf:type
rdf:type
dcterms: description dbProp: abstract
rdfs: subPropertyOf
Note: in reality, dbpedia: abstract is no subject of
Bernhard Haslhofer, Linked Data Tutorial 63 dcterms; description
Friday, March 6, 2009
RDF Schema (RDFS)
• domain and range relationships rdfs: Class
rdf: type
xyz: Article
rdfs: domain
dbProp: abstract
rdfs: range xsd: string
Bernhard Haslhofer, Linked Data Tutorial 64
Friday, March 6, 2009
RDF Schema (RDFS)
• comments: rdfs: comment
• human-readable names for resources: rdfs: label
• machine-readable information: rdfs: seeAlso, rdfs:
isDefinedBy
Bernhard Haslhofer, Linked Data Tutorial 65
Friday, March 6, 2009
RDF Schema (RDFS)
• Shortcomings:
• no cardinality constraints on properties
• no inverse / transitive / symmetric properties
• no union / disjoint / enumeration classes
Bernhard Haslhofer, Linked Data Tutorial 66
Friday, March 6, 2009
The Semantic Web Layer Cake
User Interface & Applications
Trust
Proof
Unifying Logic
Vocabulary: Ontology: Rules:
SKOS OWL RIF
Query:
SPARQL
Crypto
RDF-S
Data Interchange: RDF
XML
URI Unicode
Bernhard Haslhofer, Linked Data Tutorial 67
Friday, March 6, 2009
Web Ontology Language (OWL)
• An extension of RDFS
• logic based; provides automated reasoning support
• Three “sub-languages”
• OWL Full: maximum expressiveness and syntactic
freedom
• OWL DL: restricted to first order logic; decidable
• OWL Lite: easy to implement; provides mechanisms for
creating classification hierarchies with simple constraints
Bernhard Haslhofer, Linked Data Tutorial 68
Friday, March 6, 2009
Web Ontology Language (OWL)
• owl:Class defines a group of individuals that belong
together because of shared properties
• owl:Thing is the class of all individuals, and a superclass
of all OWL classes
rdfs: Class
rdfs: subClassOf
owl: Class
owl: Thing
rdfs: subClassOf
rdf:type
rdf:type
http://dbpedia.org/resource/
dbpedia-owl: Place
Hallstatt
Bernhard Haslhofer, Linked Data Tutorial 69
Friday, March 6, 2009
Web Ontology Language (OWL)
• DatatypeProperty: properties whose value is a literal
• ObjectProperty: properties whose value is an individual
owl:
rdf:type rdfs: range xsd: string
DatatypeProperty
dbPedia-owl: id
rdfs: subClassOf
rdfs: domain dbPedia-owl:
rdf: Property
WorldHeritageSite
dbPedia-owl:
rdf:type rdfs: domain
rdfs: subClassOf region
owl: dbpedia-owl:
rdfs: range
ObjectProperty PopulatedPlace
Bernhard Haslhofer, Linked Data Tutorial 70
Friday, March 6, 2009
Web Ontology Language (OWL)
• owl: equivalentClass states that two classes are
equivalent, i.e., they have the same instances
owl: equivalentClass
dbpedia: Car dbpedia: Automobile
Note: DBpedia actually uses dbpedia: redirect
Bernhard Haslhofer, Linked Data Tutorial 71
Friday, March 6, 2009
Web Ontology Language (OWL)
• owl: equivalentProperty states that two properties are
equivalent; they relate one individual to the same set
of other individuals
owl: equivalentProperty
myOnt: abstract dbpprop: abstract
Bernhard Haslhofer, Linked Data Tutorial 72
Friday, March 6, 2009
Web Ontology Language (OWL)
• owl:sameAs states that two individuals are the same
owl: sameAs
http://dbpedia.org/resource/
fbase: Hallstatt culture
Hallstatt_culture
Bernhard Haslhofer, Linked Data Tutorial 73
Friday, March 6, 2009
Web Ontology Language (OWL)
• owl: inverseOf (e.g., hasChild inverseOf hasParent)
• owl: transitiveProperty (e.g., hasAncestor)
• owl: symmetricProperty (e.g., knows)
• owl: functionalProperty (e.g. hasPrimaryEmployer)
• owl: inverseFunctionalProperty (e.g.,
hasSocialSecurityNumber)
Bernhard Haslhofer, Linked Data Tutorial 74
Friday, March 6, 2009
Web Ontology Language (OWL)
• owl: allValuesFrom restricts properties to values of
classes
• owl: minCardinality restricts a property’s minimum
cardinality with respect to a class
• owl: maxCardinalty restricts a property’s maximum
cardinality with respect to a class
• owl: cardinality = shortcut for min/maxCardinality
Bernhard Haslhofer, Linked Data Tutorial 75
Friday, March 6, 2009
Web Ontology Language (OWL)
• owl: intersectionOf (e.g., EmployedPerson = Person ∩
CorporateResources)
• owl: unionOf (e.g., NorthAmericanCitizen = USCitizen
∪ CanadaCitizen)
• owl: complementOf (e.g., JuniorResearcher =
Researcher - SeniorResearcher)
• owl: disjointWith (e.g., Woman ∩ Man = ∅)
Bernhard Haslhofer, Linked Data Tutorial 76
Friday, March 6, 2009
The Semantic Web Layer Cake
User Interface & Applications
Trust
Proof
Unifying Logic
Vocabulary: Ontology: Rules:
SKOS OWL RIF
Query:
SPARQL
Crypto
RDF-S
Data Interchange: RDF
XML
URI Unicode
Bernhard Haslhofer, Linked Data Tutorial 77
Friday, March 6, 2009
Simple Knowledge Organization System (SKOS)
• a family of formal languages for the definition and
representation of controlled vocabularies on the Web
• thesauri
• classification schemes
• taxonomies
• subject-heading systems
• an application of RDF/S & OWL
Bernhard Haslhofer, Linked Data Tutorial 78
Friday, March 6, 2009
Simple Knowledge Organization System (SKOS)
• vocabulary terms are represented as concepts
identified by URIs
skos: Concept
rdf: type rdf: type
dbpedia: Category:
xyz: HallstattCulture
Iron_Age_Europe
Bernhard Haslhofer, Linked Data Tutorial 79
Friday, March 6, 2009
Simple Knowledge Organization System (SKOS)
• concepts have preferred, alternate & hidden labels
• label is either preferred or alternate or hidden; one
per language
“Hallstatt Period@en”
bel
a
refL
:p
kos
s l
fLabe “Hallstatt Kultur@de”
s: pre
sko
xyz: HallstattCulture
skos:
altLab
el
“Hallstatt Culture@en”
Bernhard Haslhofer, Linked Data Tutorial 80
Friday, March 6, 2009
Simple Knowledge Organization System (SKOS)
• concepts can be organized in hierarchies by (the
inverse) skos: broader or skos: narrower relationships
dbpedia: Category:
dbpedia: Category: Iron_Age
Prehistoric_Europe
skos: broader skos: broader
skos: narrower
dbpedia: Category:
xyz: HallstattCulture
Iron_Age_Europe
Note: in DBPedia this narrower rel. does not exist
Bernhard Haslhofer, Linked Data Tutorial 81
Friday, March 6, 2009
Simple Knowledge Organization System (SKOS)
• concepts can be associated by non-hierarchical (non-
transitive) skos:related properties
skos: related
xyz: HallstattCulture xyz: HallstattCivilization
Bernhard Haslhofer, Linked Data Tutorial 82
Friday, March 6, 2009
Simple Knowledge Organization System (SKOS)
• concepts can have notes for general documentation
purposes
• scopeNote: an indication how the use of a concept is
limited in indexing practice
• definition: an explanation of the intended meaning
• example: supplies an example of the use of a concept
• historyNote: describes significant changes to the meaning
of a concept
Bernhard Haslhofer, Linked Data Tutorial 83
Friday, March 6, 2009
Simple Knowledge Organization System (SKOS)
• concepts can be defined as part of well-defined
concept schemes (thesauri, classification schemes)
dc:title
“Archeology Thesaurus”
ex: archeologyThesaurus
dc:creator
“John Doe”
skos: inScheme
skos: inScheme
xyz: HallstattCulture xyz: HallstattCivilization
Bernhard Haslhofer, Linked Data Tutorial 84
Friday, March 6, 2009
Simple Knowledge Organization System (SKOS)
• Other features
• SKOS allows the mapping between different concept
schemes
• concepts can be organized in collections
• relationships between concept labels
• definition of broaderTransitive and narrowerTransitive
relationships
• ....
Bernhard Haslhofer, Linked Data Tutorial 85
Friday, March 6, 2009
The Semantic Web Layer Cake
User Interface & Applications
Trust
Proof
Unifying Logic
Vocabulary: Ontology: Rules:
SKOS OWL RIF
Query:
SPARQL
Crypto
RDF-S
Data Interchange: RDF
XML
URI Unicode
Bernhard Haslhofer, Linked Data Tutorial 86
Friday, March 6, 2009
SPARQL
• is a query language that allows to access RDF data
• currently read-only; no update or delete
• based on matching graph patterns
• is a protocol that defines how queries and results can
be transported over a network (over the Web)
Bernhard Haslhofer, Linked Data Tutorial 87
Friday, March 6, 2009
SPARQL
• a simple query example
SELECT ?uri
WHERE {
?uri skos:subject <http://dbpedia.org/resource/Category:Iron_Age_Europe>.
}
uri
...
dbpedia: Hallstatt_culture
...
Bernhard Haslhofer, Linked Data Tutorial 88
Friday, March 6, 2009
SPARQL
• querying multiple variables
SELECT ?uri ?abstract
WHERE {
?uri skos:subject <http://dbpedia.org/resource/Category:Iron_Age_Europe>.
?uri dbpprop:abstract ?abstract .
}
uri abstract
...
dbpedia: Hallstatt_culture The Hallstatt culture was the...
dbpedia: Hallstatt_culture Гальшта́тская культу́ра ...
Bernhard Haslhofer, Linked Data Tutorial 89
Friday, March 6, 2009
SPARQL
• querying and filtering literal values
SELECT ?uri ?label
WHERE {
?uri rdfs:label ?label .
?uri skos:subject <http://dbpedia.org/resource/Category:Iron_Age_Europe> .
FILTER regex(?label, “culture”, “i”)
}
uri label
dbpedia: Basarabi_culture Basarabi Culture
... ...
dbpedia: Hallstatt_culture Hallstatt Culture
Bernhard Haslhofer, Linked Data Tutorial 90
Friday, March 6, 2009
SPARQL
• optional graph patterns
SELECT ?uri ?label ?image
WHERE {
?uri rdfs:label ?label .
?uri skos:subject <http://dbpedia.org/resource/Category:Iron_Age_Europe> .
OPTIONAL { ?uri foaf:img ?image }.
}
Bernhard Haslhofer, Linked Data Tutorial 91
Friday, March 6, 2009
SPARQL
• optional graph patterns with constraints
SELECT ?uri ?label ?image
WHERE {
?uri rdfs:label ?label .
?uri skos:subject <http://dbpedia.org/resource/Category:Iron_Age_Europe> .
OPTIONAL {
?uri foaf:img ?image .
FILTER regex(?image, “jpg”, “i”) .
}.
}
Bernhard Haslhofer, Linked Data Tutorial 92
Friday, March 6, 2009
SPARQL
• matching alternatives
SELECT ?uri ?classification
WHERE {
?uri skos:subject <http://dbpedia.org/resource/Category:Iron_Age_Europe> .
{ ?uri rdf:type ?classification . }
UNION
{ ?uri skos:subject ?classification . }
}
Bernhard Haslhofer, Linked Data Tutorial 93
Friday, March 6, 2009
SPARQL
• solution modifiers
SELECT ?uri ?name
WHERE {
?uri skos:subject <http://dbpedia.org/resource/Category:Iron_Age_Europe> .
?uri rdfs:label ?name
}
ORDER BY ?name
LIMIT 10
OFFSET 5
Bernhard Haslhofer, Linked Data Tutorial 94
Friday, March 6, 2009
SPARQL
• Other SPARQL features:
• rich set of FILTER expressions (see xquery functions)
• CONSTRUCT, DESCRIBE, ASK queries
• Named Graphs
Bernhard Haslhofer, Linked Data Tutorial 95
Friday, March 6, 2009
The Semantic Web Layer Cake
User Interface & Applications
Trust
Proof
Unifying Logic
Vocabulary: Ontology: Rules:
SKOS OWL RIF
Query:
SPARQL
Crypto
RDF-S
Data Interchange: RDF
XML
URI Unicode
Bernhard Haslhofer, Linked Data Tutorial 96
Friday, March 6, 2009
References
• General
• T. Berners-Lee, J. Hendler, O. Lassila: The Semantic Web. Scientific
American, May 2001.
• URI
• T. Berners-Lee, R. Fielding, L. Masinter: Uniform Resource Identifier (URI):
Generic Syntax (RFC 3986), January 2005.
• M. Duerst, M. Suignard: Internationalized Resource Identifiers (IRIs) (RFC
3987), January 2005.
• RDF/S
• F. Manola, E. Miller: RDF Primer. W3C Recommendation, February 2004.
• D. Brickley, R.V. Guha: RDF Vocabulary Description Language 1.0: RDF
Schema. W3C Recommendation, February 2004.
Bernhard Haslhofer, Linked Data Tutorial 97
Friday, March 6, 2009
References
• OWL
• Deborah L. McGuiness, Frank van Harmelen (eds.): OWL Web Ontology
Language Overview. W3C Recommendation 10 February 2004. Available
at http://www.w3.org/TR/owl-features/
• Nataly F. Noy, Deborah L. McGuinness: Ontology Development 101: A
Guide to Creating Your First Ontology. Available at http://www-
ksl.stanford.edu/people/dlm/papers/ontology-tutorial-noy-mcguinness.pdf
• SKOS
• SKOS Primer: http://www.w3.org/TR/2008/WD-skos-primer-20080829/
• SPARQL
• Eric Prud’hommeaux, Andy Seaborne (eds.): SPARQL Query Language for
RDF. W3C Candidate Recommendation 14 June 2007, available at http://
www.w3.org/TR/rdf-sparql-query
Bernhard Haslhofer, Linked Data Tutorial 98
Friday, March 6, 2009
Contents
• Motivation for Linked Open Data (LOD)
• Base Technologies
• The Linked Data Principles
• Publishing Solutions and Tools
• Vocabularies & Interoperability Issues
• Demos + Discussion
Bernhard Haslhofer, Linked Data Tutorial 99
Friday, March 6, 2009
The 4 principles
• Use URIs as names for things
• Use HTTP URIs so that people can look up those
names
• When someone looks up a URI, provide useful
information
• Include links to other other URIs, so that they can
discover more things
Bernhard Haslhofer, Linked Data Tutorial 100
Friday, March 6, 2009
Use HTTP URIs....
http://dbpedia.org/resource/
Hallstatt
http://dbpedia.org/resource/
Hallstatt_culture
http://dbpedia.org/resource/
Category:World_Heritage_Sites_in_Austria
Bernhard Haslhofer, Linked Data Tutorial 101
Friday, March 6, 2009
Provide useful information ... for humans
Bernhard Haslhofer, Linked Data Tutorial 102
Friday, March 6, 2009
Provide useful information ... for machines
<http://dbpedia.org/resource/Hallstatt_culture >
p:abstract “Die Hallstattzeit bezeichnet...” @de
p:abstract “The Hallstatt culture was...” @en
rdfs:label “Hallstatt culture”@en;
rdfs:label “Hallstattzeit” @de;
rdfs:label “Civilisation de Hallstatt” @fr;
rdfs:label “ ” @jp;
skos:subject dbpedia:Category:Iron_Age_Europe;
...
Bernhard Haslhofer, Linked Data Tutorial 103
Friday, March 6, 2009
Provide useful information for humans and machines
• return different HTTP responses depending on HTTP-
Accept-Header
Bernhard Haslhofer, Linked Data Tutorial 104
Friday, March 6, 2009
Provide useful information for humans and machines
• best practice: also assign names (URIs) to various
representations
http://dbpedia.org/resource/
Hallstatt_culture
Accept: application/rdf+xml Accept: text/html
http://dbpedia.org/data/ http://dbpedia.org/page/
Hallstatt_culture.rdf Hallstatt_culture
Bernhard Haslhofer, Linked Data Tutorial 105
Friday, March 6, 2009
Include links
dbpprop: hasPhotoCollection http://www4.wiwiss.fu-berlin.de/
flickrwrappr/photos/Hallstatt_culture
http://dbpedia.org/resource/
Hallstatt_culture
owl: sameAs
http://rdf.freebase.com/ns/guid.
9202a8c04000641f8000000000239c44
Bernhard Haslhofer, Linked Data Tutorial 106
Friday, March 6, 2009
Include links
• other common properties
• rdfs: seeAlso
• foaf: knows
• foaf: based_near
• foaf: topic_interest
Bernhard Haslhofer, Linked Data Tutorial 107
Friday, March 6, 2009
Include links
• manual link generation
• only for small data sets
• automatic link generation
• pattern-based algorithms (e.g., same ISBN number)
• more complex property-based algorithms
• see record-linkage problem in database domain
Bernhard Haslhofer, Linked Data Tutorial 108
Friday, March 6, 2009
Benefit
• Clients can easily look up names and retrieve
information and follow the links
Bernhard Haslhofer, Linked Data Tutorial 109
Friday, March 6, 2009
References
• Berners-Lee 2006: Linked Data. Available at http://
www.w3.org/DesignIssues/LinkedData.html
• Bizer et al.: How to Publish Linked Data on the Web.
Available at: http://www4.wiwiss.fu-berlin.de/bizer/pub/
LinkedDataTutorial/
Bernhard Haslhofer, Linked Data Tutorial 110
Friday, March 6, 2009
Contents
• Motivation for Linked Open Data (LOD)
• Base Technologies
• The Linked Data Principles
• Publishing Solutions and Tools
• Vocabularies & Interoperability Issues
• Demos + Discussion
Bernhard Haslhofer, Linked Data Tutorial 111
Friday, March 6, 2009
Relational Databases
Triplify D2R Server Virtuoso RDF Views
Scripting languages
Java Middleware Solution
Technology (PHP)
- Yes Yes
SPARQL Endpoint
SQL RDF based RDF based
Mapping Language
Manual Semi-automatic Manual
Mapping Generation
Medium-high
medium high
Scalability (but no SPARQL)
(c) Sören Auer, http://www.slideshare.net/soeren1611/linked-data-
Bernhard Haslhofer, Linked Data Tutorial 112 tutorial-presentation-955375
Friday, March 6, 2009
Triplify
• Goal: expose semantics available in RDBMS as simple
as possible
• Available for most popular Web app languages
• PHP (ready), Ruby/Python (under dev.)
• Works with most popular Web app databases
• MySQL, PHP-PDO DBs (SQLite, Oracle, DB2, MS SQL,
PostgreSQL)
Bernhard Haslhofer, Linked Data Tutorial 113
Friday, March 6, 2009
Triplify
• Configuration
• number of SQL queries selecting information, which
should be made publicly available
SELECT id, name AS ‘foaf:name’ FROM users
Bernhard Haslhofer, Linked Data Tutorial 114
Friday, March 6, 2009
Triplify
!quot;quot;#$%&'()'*()+,-).&'*)/0&12(3')4.#quot;'5)6%'*)&)3728(0)#/)
9:-).&;(03quot;<)
!quot;#$%%&'()*+,-.*(/)%0/1#'123%#(-0%455567
SELECT id, post_author AS 'sioc:has_creator->user',
post_title AS 'dc:title',
post_content AS 'sioc:content', post_date AS
'dcterms:modified^^xsd:dateTime‘,
post_modified AS 'dcterms:created^^xsd:dateTime'
FROM posts
WHERE post_status='publish‘ (AND id=xxx)
SELECT post_id id, tag_label AS 'tag:taggedWithTag‘
FROM post2tag INNER JOIN tag ON(post2tag.tag_id=tag.tag_id)
(WHERE id=xxx)
SELECT post_id id, category_id AS 'belongsToCategory->category‘
FROM post2cat
(WHERE id=xxx)
Bernhard Haslhofer, Linked Data Tutorial 115
Friday, March 6, 2009
D2R Server
• a tool for publishing the data from relational databases
on the Web
Bernhard Haslhofer, Linked Data Tutorial 116
Friday, March 6, 2009
D2R Server
• D2RQ mapping excerpt
map:Conference a d2rq:ClassMap;
d2rq:dataStorage map:Database1.
d2rq:class :Conference;
d2rq:uriPattern quot;http://conferences.org/comp/confno@@Conferences.ConfID@@quot;;
.
map:eventTitle a d2rq:PropertyBridge;
d2rq:belongsToClassMap map:Conference;
d2rq:property :eventTitle;
d2rq:column quot;Conferences.Namequot;;
d2rq:datatype xsd:string;
.
Bernhard Haslhofer, Linked Data Tutorial 117
Friday, March 6, 2009
D2R Server
• supports auto-generation of mapping files
• extracted from database structure
• good starting point
• requires adaption for domain-specific vocabulary
• supports dumping databases to RDF files
Bernhard Haslhofer, Linked Data Tutorial 118
Friday, March 6, 2009
Virtuoso RDF Views
• transforms the result of SQL SELECT statements into
RDF
• mapping steps
• define RDFS class IRIs for each table
• define construction of subject IRIs from primary key
column values
• define construction of predicate IRIs from each non-key
column
Bernhard Haslhofer, Linked Data Tutorial 119
Friday, March 6, 2009
OAI2LOD Server
• publishes metadata from arbitrary OAI-PMH
endpoints as linked data on the Web
• provides a simple linking framework
Bernhard Haslhofer, Linked Data Tutorial 120
Friday, March 6, 2009
OAI2LOD Server
• Sample harvesting configuration:
<> a oai2lod:Server;
rdfs:label quot;Example OAI2LOD Serverquot;;
oai2lod:port 2020;
oai2lod:baseURI <http://localhost:2020/>;
oai2lod:publishes <oai1>;
oai2lod:linkedWith <link1>;
.
<oai1> a oai2lod:OAIServer;
oai2lod:serverURL <http://oai-bdb.onb.ac.at/Script/oai2.aspx>;
oai2lod:metadataPrefix quot;oai_dcquot;;
oai2lod:styleSheet quot;xsl/oai_dc2rdf_xml.xslquot;;
oai2lod:maxRecords 50;
.
Bernhard Haslhofer, Linked Data Tutorial 121
Friday, March 6, 2009
OAI2LOD Server
• Sample linking configuration:
<link1> a oai2lod:LinkedSPARQLEndpoint;
oai2lod:sparqlService <http://DBpedia.org/sparql>;
oai2lod:maxResults 5000;
oai2lod:linkingRule <lrule1>;
.
<lrule1> a oai2lod:LinkingRule;
oai2lod:sourceType <http://www.mediaspaces.info/vocab/oai-pmh.rdf#Item>;
oai2lod:sourceProperty <http://purl.org/dc/elements/1.1/subject>;
oai2lod:targetType <http://dbpedia.org/class/yago/Capital108518505>;
oai2lod:targetProperty <http://www.w3.org/2000/01/rdf-schema#label>;
oai2lod:linkingProperty <http://www.w3.org/2000/01/rdf-schema#seeAlso>;
oai2lod:similarityMetrics quot;uk.ac.shef.wit.simmetrics.similaritymetrics.Levenshteinquot;;
oai2lod:minSimilarity 1.0;
.
Bernhard Haslhofer, Linked Data Tutorial 122
Friday, March 6, 2009
SILK - Link Discovery Framework
• Supports data publishers in linking their data sets with
other
• Provides a declarative Link Specification Language
• specify which links to be discovered
• specify conditions data items must fulfill in order to be
linked
• implemented in Python
Bernhard Haslhofer, Linked Data Tutorial 123
Friday, March 6, 2009
References
• A Survey of current approaches for mapping of relational databases to RDF:
http://esw.w3.org/topic/Rdb2RdfXG/StateOfTheArt
• Triplify: http://triplify.org/Overview
• D2R Server: http://www4.wiwiss.fu-berlin.de/bizer/d2r-server/
• OpenLink Virtuoso: http://virtuoso.openlinksw.com/Whitepapers/html/
rdf_views/virtuoso_rdf_views_example.html
• OAI2LOD Server: http://www.mediaspaces.info/tools/oai2lod
• SILK: http://www4.wiwiss.fu-berlin.de/bizer/silk/
Bernhard Haslhofer, Linked Data Tutorial 125
Friday, March 6, 2009
Contents
• Motivation for Linked Open Data (LOD)
• Base Technologies
• The Linked Data Principles
• Publishing Solutions and Tools
• Vocabularies & Interoperability Issues
• Demos + Discussion
Bernhard Haslhofer, Linked Data Tutorial 126
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• LOD is about publishing data on the Web
• vocabularies define the terms that describe the
semantics of these data (e.g., creator, title, abstract,
etc.)
• we can apply the LOD principles and expose
vocabularies on the Web
Bernhard Haslhofer, Linked Data Tutorial 127
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• Use URIs as names for vocabularies and terms
• Use HTTP URIs so that people (and machines) can
look up those names
• When someone looks up a URI, provide useful
information
• Include links (mappings) to other vocabularies
Bernhard Haslhofer, Linked Data Tutorial 128
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• Example: Dublin Core terms
Bernhard Haslhofer, Linked Data Tutorial 129
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• Example: Dublin Core terms
Bernhard Haslhofer, Linked Data Tutorial 130
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• Example: Friend of a Friend (FOAF)
Bernhard Haslhofer, Linked Data Tutorial 131
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• Example: Friend of a Friend (FOAF)
Bernhard Haslhofer, Linked Data Tutorial 132
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• Publishing vocabularies on the Web recipe
Bernhard Haslhofer, Linked Data Tutorial 133
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• Publishing vocabularies on the Web recipe
• Step 1: create a complete RDF/XML serialization of the
vocabulary (e.g., example.rdf)
• Step 2: copy the serialization into Web Server directory
• Step3: add a .htaccess directive to the Web Server
directory
Bernhard Haslhofer, Linked Data Tutorial 134
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• Publishing vocabularies on the Web recipe
Bernhard Haslhofer, Linked Data Tutorial 135
Friday, March 6, 2009
Vocabularies & Interoperability Issues
• LOD and the Semantic Web provide the technologies
to expose data on the Web
• exposing data in RDF and describing vocabularies in
RDF/S or OWL provides data interoperability on a
technical / structural level
• semantic interoperability is not given by default
Bernhard Haslhofer, Linked Data Tutorial 136
Friday, March 6, 2009
Vocabularies & Interoperability Issues
rule-of-thumb to establish semantic interoperability
1. whenever possible, use terms from widely used vocabularies that are
published on the Web in a structured, machine-readable format (RDFS
or OWL)
2. if there is no such term that reflects the required semantics, re-use a
semantically broader term by establishing e.g. a rdfs:subProperty
relationship, refine its semantics for the purpose of your application
within a new namespace, and publish it on the Web
3. if (1) and (2) are not feasible, create your own vocabulary and publish
it on the Web in order to make it accessible for other users and
applications; consider to define mappings to other vocabularies
Bernhard Haslhofer, Linked Data Tutorial 137
Friday, March 6, 2009
Vocabularies & Interoperability Issues
Concepts/
Name Purpose Base
Properties
Dublin Core General /
RDFS 22 / 55
(http://purl.org/dc/terms) Documents
Friend of a Friend (FOAF) Contacts /
OWL 12 / 54
(http://xmlns.com/foaf/0.1/) Communication
VCard Ontology Contacts /
OWL 5 / 54
(http://www.w3.org/2006/vcard/ns#) Communication
Description of a Project (DOAP)
Projects RDFS 7 / 30
(http://usefulinc.com/ns/doap# )
Bernhard Haslhofer, Linked Data Tutorial 138
Friday, March 6, 2009
Vocabularies & Interoperability Issues
Concepts/
Name Purpose Base
Properties
Semantically Interlinked Online Communities
Contacts /
(SIOC) OWL 11 / 53
Communication
(http://rdfs.org/sioc/ns#)
Nepomuk Message Ontology (NMO)
Contacts /
(http://www.semanticdesktop.org/ontologies/ RDFS 7 / 23
Communication
2007/03/22/nmo# )
GeoNames Ontology
Locations OWL 7 / 18
(http://www.geonames.org/ontology#)
Music Ontology
Multimedia OWL 53 / 131
(http://purl.org/ontology/mo/ )
Bernhard Haslhofer, Linked Data Tutorial 139
Friday, March 6, 2009
References
• Miles et al.: Best Practices Recipes for Publishing RDF Vocabularies, Available
at: http://www.w3.org/TR/swbp-vocab-pub/
• Heery and Patel 2004: Application profiles: mixing and matching metadata
schemes. Available at: http://www.ariadne.ac.uk/issue25/app-profiles/
• Haslhofer and Klas 2009: A Survey of approaches for achieving metadata
interoperability. To be published in ACM Computing Surveys Q4/2009.
Available on request.
• Haslhofer and Schandl 2009: Interweaving OAI-PMH metadata with the
Linked Data Cloud. To be published in International Journal of Metadata,
Semantics, and Ontologies. Available on request.
Bernhard Haslhofer, Linked Data Tutorial 141
Friday, March 6, 2009
Contents
• Motivation for Linked Open Data (LOD)
• Base Technologies
• The Linked Data Principles
• Publishing Solutions and Tools
• Vocabularies & Interoperability Issues
• Demos + Discussion
Bernhard Haslhofer, Linked Data Tutorial 142
Friday, March 6, 2009
Demo 1 - browsing and querying DBPedia
• http://dbpedia.org/resource/Hallstatt_culture
• http://dbpedia.org/snorql
• http://lookup.dbpedia.org
Bernhard Haslhofer, Linked Data Tutorial 143
Friday, March 6, 2009
Demo 2 - cURLing the LIBRIS catalogue
• LIBRIS = The Swedish Union Catalog
• Starting point: http://libris.kb.se/bib/10432900
Bernhard Haslhofer, Linked Data Tutorial 144
Friday, March 6, 2009
Demo 3 - exposing MySQL DB using D2R
• sample database: http://dev.mysql.com/doc/world-
setup/en/world-setup.html
Bernhard Haslhofer, Linked Data Tutorial 145
Friday, March 6, 2009
Demo 4 - sample OAI2LOD instances
• http://www.mediaspaces.info/tools/oai2lod/
Bernhard Haslhofer, Linked Data Tutorial 146
Friday, March 6, 2009
Demo 5 - RDF Viewers
• Zitgist: http://dataviewer.zitgist.com/
• Disco: http://www4.wiwiss.fu-berlin.de/rdf_browser/
• OpenLink Data Explorer: http://demo.openlinksw.com/
rdfbrowser2
Bernhard Haslhofer, Linked Data Tutorial 147
Friday, March 6, 2009
Known / Open Issues
• Automatic linking is non-trival and domain specific
• how to deal with false positives?
• precision / recall of existing approaches?
• Nobody can guarantee the (long-term) availability of
LOD resources
• what to do when a resource disappears?
• annoying for humans / problematic for applications
• Licensing
Bernhard Haslhofer, Linked Data Tutorial 148
Friday, March 6, 2009
Summary
• Linked Data is about publishing and interlinking public
interest data on the Web
• Other applications can access these data with common
Web technologies
• It is still research, with lots of construction areas, but
within its two year (!!!) history it has attracted quite a lot
of interest
• It is an exciting research field to work on...
Bernhard Haslhofer, Linked Data Tutorial 149
Friday, March 6, 2009