Linked Data and RDF Integrating information sources Inference - PowerPoint PPT Presentation

Building a Semantic Web Annotation • – Associating metadata with resources Integration • Linked Data and RDF – Integrating information sources Inference • – Reasoning over the information we have. – Could be light-weight (taxonomy) – Could be heavy-weight (logic-style) COMP60421 Interoperation and Sharing are key goals • Sean Bechhofer sean.bechhofer@manchester.ac.uk 6

Linked Data* Database tables Linked Data or the Data Web is about using the Web to connect • isbn title author publisherId pages related data that wasn’t previously linked. The intention is that we move from a web of documents to a web of • 0743267478 Q&A Vikas Swarup 1435 336 data – The Web as database 014029466X The Rotters’ Jonathan Coe 1546 416 Club The Linked Data approach builds heavily on RDF. • … … … … … .. … … … … *Linked data slides based on material from Ian Davis and Tom Heath: http://www.slideshare.net/iandavis/30-minute-guide-to-rdf-and-linked-data 8 9

Columns represent Rows represent “things” “properties” isbn title author publisherId pages isbn title author publisherId pages 0743267478 Q&A Vikas Swarup 1435 336 0743267478 Q&A Vikas Swarup 1435 336 014029466X The Rotters’ Jonathan Coe 1546 416 014029466X The Rotters’ Jonathan Coe 1546 416 Club Club … … … … … … … … … … .. … … … … .. … … … … 10 11

Intersections represent Graphical Representation properties of things isbn title author publisherId pages title book The Rotters’ Club 0743267478 Q&A Vikas Swarup 1435 336 014029466X The Rotters’ Jonathan Coe 1546 416 Club more generally: … … … … … .. … … … … property subject value 12 13

Selecting multiple properties Multiple properties graphically isbn title author publisherId pages 0743267478 Q&A Vikas Swarup 1435 336 title The Rotters’ Club 014029466X The Rotters’ Jonathan Coe 1546 416 isbn book 014029466X Club … … … … … author Jonathan Coe .. … … … … 14 15

Relations between “things” Identification We need to be able to identify things globally and uniquely. • URIs provide this capability • Key to Linked Data is the use of URIs, specifically http:// URIs. • title The Rotters’ Club isbn book 014029466X author Jonathan Coe publisher name publisher Penguin Books 16 17

URIs in graphs URIs and naming URIs identify the things we are describing. • If two people create data using the same URI, the assumption is that URIs as names for nodes • they are describing the same thing. Merging/integrating data then becomes easy • – Although introduces issues of URI control. http://example.com/name http://example.com/person/176 Jonathan Coe URIs as names for relations 18 19

Graph Merging Graph Merging http://example.com/person/176 http://example.com/person/176 http://example.com/author http://example.com/author http://example.com/name http://example.com/name http://example.com/book/ 014029466X http://example.com/book/ 014029466X Jonathan Coe Jonathan Coe http://example.com/birthplace http://example.com/person/176 http://example.com/place/xyz765 http://example.com/birthplace http://example.com/name http://example.com/place/xyz765 http://example.com/name Birmingham Birmingham 20 21

URIs are active Linked Data Principles URIs can be more than just names -- they can be dereferenced, and 1. Use URIs as names for things • information can be retrieved. 2. Use http URIs so that those names can be dereferenced. In particular, we can lookup the URIs in a graph and potentially • 3. When a URI is looked up, provide useful information retrieve more information about the URI. 4. Include statements that link to other URIs so that more information “Follow your nose” navigation • can be discovered. Information should be returned in appropriate, machine readable • formats (e.g. another graph) Common infrastructure facilitates construction of applications. • – Largely browsers up to now…. Other guidelines relating to connecting documents with the data that • describes them. – Use of content negotiation to supply “appropriate” representations – Use of microformats/RDFa to publish data 22 23

RDF The RDF Data Model RDF stands for Resource Description Framework Statements are <subject, predicate, object> triples: • • It is a W3C Recommendation – <Sean,hasColleague,Uli> • – http://www.w3.org/RDF Can be represented as a graph: • RDF is a graphical formalism ( + concrete syntax) • hasColleague Sean Uli – for representing metadata – for describing the semantics of information in a machine- Statements describe properties of resources • accessible way A resource is any object that can be pointed to by a URI • Provides a simple data model based on triples. • Properties themselves are also resources (URIs) • Allows us to represent relationships between things. • 24 25

Linking Statements RDF Syntax The subject of one statement can be the object of another RDF has an XML syntax that has a specific meaning: • • Such collections of statements form a directed, labeled graph Every Description element describes a resource • • Every attribute or nested element inside a Description is a • property of that Resource “Sean K. Bechhofer” We can refer to resources by URIs • hasName hasColleague Sean Uli hasHomePage <Description about="some.uri/person/sean_bechhofer"> <hasColleague resource="some.uri/person/uli_sattler"/> hasColleague <hasName rdf:datatype="&xsd;string">Sean K. Bechhofer</hasName> </Description> Carole http://www.cs.man.ac.uk/~sattler <Description about="some.uri/person/uli_sattler"> <o:hasHomePage>http://www.cs.mam.ac.uk/~sattler</o:hasHomePage> </Description> Note that the object of a triple can also be a “literal” (a string) • <Description about="some.uri/person/carole_goble"> <o:hasColleague resource="some.uri/person/uli_sattler"/> </Description> 26 27

What does RDF give us? Querying RDF: SPARQL RDF provides us with a way of representing information as a graph A mechanism for annotating data and resources. • • SPARQL allows us to query this information Single (simple) data model. • • http://www.w3.org/TR/sparql11-overview/ Syntactic consistency between names (URIs). • Low level integration of data. Provides a query language and the description of a protocol for • • interacting with SPARQL “endpoints” via HTTP The Linked Data/Web of Data approach. • PREFIX etree:<http://etree.linkedmusic.org/vocab/> PREFIX mo:<http://purl.org/ontology/mo/> PREFIX event:<http://purl.org/NET/c4dm/event.owl#> PREFIX skos:<http://www.w3.org/2004/02/skos/core#> PREFIX timeline:<http://purl.org/NET/c4dm/timeline.owl#> PREFIX rdf:<http://www.w3.org/1999/02/22-rdf-syntax-ns#> SELECT DISTINCT ?artist WHERE { ?art rdf:type mo:MusicArtist. ?art skos:prefLabel ?artist. } 28 24

RDF(S): RDF Schema RDF(S) RDF gives a formalism for meta data annotation, and a way to write it These terms are the RDF Schema building blocks (constructors) used • • to create vocabularies: down in XML, but it doesn’t give any special meaning to vocabulary such as subClassOf or type – <Person,type,Class> – <hasColleague,type,Property> – Interpretation is an arbitrary binary relation – <Professor,subClassOf,Person> RDF Schema extends RDF with a schema vocabulary that allows you • – <Carole,type,Professor> to define basic vocabulary terms and the relations between those terms – <hasColleague,range,Person> – <hasColleague,domain,Person> – Class, Property Semantics gives “extra meaning” to particular RDF predicates and – type, subClassOf • resources – range, domain – specifies how terms should be interpreted 29 30

What does RDF(S) give us? Problems with RDF(S) Ability to use simple schema/vocabularies when describing our RDF(S) is too weak to describe resources in sufficient detail • • resources. – No localised range and domain constraints Consistent vocabulary use and sharing. • § Can’t say that the range of hasChild is Person when applied to Persons and Elephant when applied to Elephants Basic inference • – No existence/cardinality constraints § Can’t say that all instances of Person have a mother that is also a Note that RDF is a data model. There are many ways of serialising this • Person, or that Persons have exactly 2 parents data: – No transitive, inverse or symmetrical properties – RDF/XML § Can’t say that isPartOf is a transitive property, that hasPart is the – Turtle inverse of isPartOf or that touches is symmetrical – N3 Difficult to provide reasoning support • – json-ld – No “native” reasoners for non-standard semantics – May be possible to reason via FO axiomatisation 34 35