Describing Resources on the Web: The Resource Description Framework

1
Describing Resources on the Web The Resource
Description Framework
Vassilis ChristophidesDimitris Plexousakis
Computer Science Department, University of
CreteInstitute for Computer Science -
FORTHHeraklion, Crete http//www.ics.forth.gr/pr
oj/isst/RDF
2
Introduction to Metadata
3
What is the Problem?

• 3.6 million Web sites
• Five hundred million or more addressable pages on
  the Web
• High consumer expectations conflicting with
  primitive tools and mechanisms
• Uncertain quality, integrity, trust

4
The Information Landscape in the Web-era

• The Web changes relationships among
• authors
• publishers
• information intermediaries and distributors
• users
• Lower barriers to publication
• rapid dissemination of information and ideas
• less advantage to size or centralization
• greatly expanded access
• Manageability is reduced
• resource discovery is chaotic
• organization is haphazard
• preservation is almost non-existent

5
The Web Information System vs. Traditional
Libraries

• Search systems are motivated by advertising
• Index coverage is unpredictable and limited (1/3)
• Too much recall, too little precision
• Index spam abound
• Resources (and their names) are volatile
• What about versions, editions, back issues?
• Archiving is presently unsolved
• Authority and quality of service are spotty
• Managing Access Rights is hard

6
Metadata Higher Quality Web Information Services

• Traditionally
• metadata has been understood as Data about Data
• help to impose order on chaos
• Example(s)
• a library catalogue contains information
  (metadata) about publications (data)
• a file system maintains permissions (metadata)
  about files (data)
• Metadata describes other data
• One applications metadata is another
  applications data
• Metadata can itself be described by metadata (but
  that doesnt make it meta-metadata)
• Example
• Price lists (metadata) have expiration dates
  metadata about metadata (It is still just
  metadata!!)

7
Metadata takes Many Forms
resource
document
rights
discovery
administration
management
content
security and
archival
rating
authentication
status
products and
database
process control
services
schemas
or description
8
Metadata exists for Almost Anything

• People
• Places
• Objects
• Concepts
• Documents
• Archives
• Databases

9
Application Item and Collection Cataloguing

• Describing individual resources
• documents, pages, images, audio files, etc.
• Describing the content of collections
• Web sites, databases, directories, etc.
• Relationships among Resources
• Tables of Content, chapters, images.
• Site Maps

10
Application Resource Discovery

• Search engines can better understand the
  contents of a particular page
• More accurate searches
• Additional information aids precision
• Makes it possible to automate searches because
  less manual weeding is needed to process the
  search results

11
Application Electronic Commerce
Broker

• Metadata can be used to encode information needed
  in all stages of electronic commerce
• locating seller/buyer product
• searching yellow pages
• agreeing on terms of sale
• prices, terms of payment, contractual information
• transactions
• delivery mechanisms, dates, terms

Market place
Providers/Clients
12
Application Intelligent Agents

• Representation and sharing of knowledge
• knowledge exchange
• modeling
• Communication
• user-to-agent, agent-to-agent, agent-to-service
• Resource discovery
• gives web-roaming agents the ability to
  understand their environment

13
Application Content Rating

• Empowering users to select which kinds of web
  content they wish to see
• Child Protection
• W3C PICS (Platform for Internet Content
  Selection) working group
• US Communications Decency Act of 1996
• simple metadata architecture
• precursor to RDF

14
Application Digital Signatures

• These are key to building the Web of Trust
• Required by
• agents
• electronic commerce
• collaboration
• RDF will become the preferred way to encode
  digital signatures on documents and on statements
  about documents

15
Other Applications

• Privacy Preferences and Policies
• describing a users willingness/ reluctance to
  disclose information about him/her-self
• describing a site administrators desire to
  gather information about visiting users
• Intellectual Property Rights
• contractual terms related to usage and
  distribution rights to a document

16
(Meta)Data Transmission Methods
17
Metadata Assertions

• The Web is machine-readable but not
  machine-understandable
• Metadata is useful
• A lot could be gained from structured description
  of pages, servers, search services, and other
  resources
• Accommodate multiple varieties of metadata
• Metadata requirements will evolve

18
A Plethora of Metadata Standards

• Many metadata standards have evolved at different
  levels, and to meet different requirements...

MICI
19
Interoperability Issues
Lets talk English
Standardisation ofcontent
cat milk sat drank mat
SemanticInteroperability
Heres how to make a sentence
Cat sat on mat. Drankmilk.
Standardisation ofform
StructuralInteroperability
These are the rulesof grammar
The cat sat on the mat.It drank some milk.
SyntacticInteroperability
Standardisation ofexpression
20
Metadata Challenges

• Many flavours of metadata
• which one do I use?
• Managing change
• new varieties, and evolution of existing forms
• Tension between functionality and simplicity,
  extensibility and interoperability

21
Towards Metadata for Community Webs

• Group of people sharing a domain of discourse and
  a set of resources (e.g., data, documents,
  services) and having some common interests
• Commerce, Education, Health
• Provide community-specific metadata functionality
  in order to create, administrate, and access
  resources
• common semantic, structural, and syntactic
  conventions for exchange of resource description
  information

Community Webs
Workplace
Education
Commerce
Health
22
Metadata Interoperability in Community Webs

• Communities of expertise (not software vendors)
  are responsible for
• Semantics
• Registration
• Administration
• Access management
• Authority of data
• Sharing and Distribution

23
Metadata Implementation Approaches

• Harvesting metadata into a repository (database)
• Distributed Database Search

24
Harvesting Metadata into a Repository (database)
25
Distributed Database Search
26
Understanding RDF
RDF
27
RDF origins

• W3C Metadata Activity 1997-2000
• PICS (Internet content selection)
• Warwick Framework / Dublin Core
• XML (XML Data, Channels etc)
• MCF (Apple, Netscape)
• URI specification for Web identifiers

28
RDF Objectives

• Enables resource description communities to
  define their own semantics
• We can disagree about semantics, but share
  infrastructure (syntax, query, editors)
• Imposes structural constraints on the expression
  of various application metadata
• for consistent encoding, exchange and processing
  of metadata on the Web
• Metadata vocabularies can be developed without
  central coordination
• Fine-grained mixing of diverse metadata
• Signed RDF is the basis for trust
• XML used for serialisation syntax

29
Describing Community Resources using RDF
30
The Basic RDF Data Model

• RDF Resource Descriptions
• Data Model Directed Labeled Graphs
• Nodes Resources (URIs) or Literals
• Edges Properties Attributes or Relationships
• Statement assertion of the form resource,
  property, value
• Description set of statements concerning a
  resource
• XML syntax

31
The Basic RDF Data Model Primitives
Resource
Property
Value
32
Simple Example
URITutorial
Author
Vassilis
33
The notion of Resource

• A resource is identified by a URI
• absoluteURI relativeURI fragment-id
• The resource identified by a URI may be abstract
• i.e. not network retrievable
• Resource is distinct from entity resolved at any
  particular time
• http//www.ics.forth.gr/RDF/
• From RFC 2396
• Resource A resource can be anything that has
  identity. Familiar examples include an electronic
  document, an image, a service (e.g., "today's
  weather report for Los Angeles"), and a
  collection of other resources. Not all resources
  are network "retrievable" e.g., human beings,
  corporations, and bound books in a library can
  also be considered resources. The resource is the
  conceptual mapping to an entity or set of
  entities, not necessarily the entity which
  corresponds to that mapping at any particular
  instance in time. Thus, a resource can remain
  constant even when its content---the entities to
  which it currently corresponds---changes over
  time, provided that the conceptual mapping is not
  changed in the process.

34
RDF Syntax

• RDF Model defines a formal relationships among
  resources, properties and values
• Syntax is required to...
• Store instances of the model into files
• Communicate files from one application to another
• W3C XML eXtensible Markup Language

lttag1gt lttag2gt lttag3gt lt/tag1gt
35
RDF Model Example Complex Values
URITutorial
Title
RDF Presentation
Creator
Vassilis Christophides
ltRDF xmlns http//www.w3.org/TR/WD-rdf-syntax
xmlnsdc http//purl.org/dc/element
s/1.0/gt ltDescription about URITutorialgt
ltdcTitlegt RDF Presentation lt/dcTitlegt
ltdcCreatorgt Vassilis Christophides
lt/dcCreatorgt lt/Descriptiongt lt/RDFgt
36
RDF Syntax Example Complex Values
ltRDF xmlns http//www.w3.org/TR/WD-rdf-syntax
xmlnsdc http//purl.org/dc/element
s/1.0/ xmlnsbib http//www.bib.org
/personsgt ltDescription about
URITutorialgt ltdcTitlegt RDF Presentation
lt/dcTitlegt ltdcCreatorgt
ltDescriptiongt ltbibNamegt Vassilis
Christophides lt/bibNamegt ltbibEmailgt
christop_at_ics.forth.gr lt/bibEmailgt
ltbibAff resource http//www.ics.forth.gr /gt
lt/Descriptiongt lt/dcCreatorgt
lt/Descriptiongt lt/RDFgt
ltDescription bibName Vassilis
Christophides bibEmail christop_at_ics.forth.g
r gt ltbibAff resource http//www.ics.forth.g
r /gt lt/Descriptiongt
37
RDF Model Example
38
Where do you stop?

• The Basic RDF model syntax provides enabling
  technology
• Degree of metadata simplicity/complexity is a
  matter of
• Resource description communities needs,
  best-practice and experience
• Organization/Institutions Policy
• Economics
• Goals and requirements of implementation

39
The Basic RDF Data Model In Brief
40
One Additional Concept Container Values

• Containers are collections
• they allow grouping of resources (or literal
  values)
• It is possible to make statements about the
  container (as a whole) or about its members
  individually
• Different types of containers exist
• Bags -- groups of things
• Sequences -- ordered group of things
• Alternates -- Alternate things/values
• First value is the default
• Must be at least one
• Duplicate values are permitted
• there is no mechanism to enforce unique value
  constraints
• Syntactic shorthand provided (much like HTML
  lists)

41
Containers (continued)
URITutorial
dcCreator
rdfType
rdfSeq
rdf_1
rdf_2
Vassilis Christophides
Dimitris Plexousakis
42
Containers (continued)
URITutorial
dcCreator
dcCreator
Vassilis Christophides
Dimitris Plexousakis
43
The Basic RDF Data Model Formal Aspects

• Statement (predicate,subject,object)
• Predicate is a resource
• Subject is a resource
• Object is either a resource or a literal
• Object Predicate(Subject)
• A model is a set of statements
• Formal model based on triples (Universal
  relation)
• Example
• author, http//www.ics.forth.gr/proj/isst/RDF,
  node
• name, node, Vassilis Christophides
• email, node, christop_at_ics.forth.gr

44
Triples for Container Values Example

• Triples from the first example
• http//www.ics.forth.gr/proj/isst/RDF,dcCreato
  r,x
• x, rdf_1, Vassilis Christophides
• x, rdf_2, Dimitris Plexousakis
• x, rdftype, rdfSeq
• Triples from the second example
• http//www.ics.forth.gr/proj/isst/RDF,dcCreato
  r, Vassilis Christophides
• http//www.w3.org/TR/REC-rdf-syntax,
  dcCreator, Dimitris Plexousakis

45
Edge Labeled Directed Graphs (RDF)
Vassilis
creator
affiliation
RDFTutorial
ICS-FORTH
projects
activities
ISL
C-Web
(creator, RDFTutorial, Vassilis) (affiliation,
Vassilis, ICS-FORTH) (activities, ICS-FORTH,
ISL) (projects, ICS-FORTH, C-Web)
46
Node labeled Directed Graph (XML)
root
x
element
element
2
attribute
foo
bar
element
attribute
attribute
attribute
baz
href
x
y
attribute
1
3
z
ltrootgt ltfoo href x1 /gt ltbar x2
y3gt ltbaz zaaa/gt lt/bargt
aaa
47
What can we Express in RDF?

• RDF relies on a (edge labeled) directed graph
  model that can easily
• extended by just adding more edges
• combine multiple vocabularies, distinguished by
  their URIs
• RDF provides a standard syntax to represent these
  graphs in XML
• RDF Model can be thought of as a simplified XML
  Infoset
• But RDF goes beyond XML syntactic issues
• It allows to define semantic networks on the Web

48
Semantic Networks
name
Person
String
lives in
isa
creates
Artist
Artifact
isa
isa
isa
isa
Painter
Sculptor
Sculpture
Painting
isa
paints
sculpts
a Person has a name and lives_in somewhere .
Artists are persons, painters and sculptors are
artists. An artist creates artifacts, (paintings
or sculptures) a painter paints paintings and a
sculptor sculpts sculptures
49
RDF Schema Definition RDFS

• Declaration of label vocabularies for description
  graph nodes edges
• Enables communities to share machine readable
  tokens and define human readable labels
• Node labels (types) are defined as classes
• Literal data types as defined by XML Schemas WG
• Resource may have a specific type property
• Edge labels (predicates) are defined as
  properties of these classes
• A resource of given type may have a given
  property (domain constraint)
• A resource of given type may be the value of a
  given predicate (range constraint)
• RDFS vocabularies expressible in the basic RDF
  model and syntax
• RDFS vocabularies are also Web resources (and
  have URIs) and therefore can be described using
  RDF

50
Constructing and Using RDF schemas

• RDFS Schema Vocabularies allows for
• Specialization of both classes properties
  (simple multiple)
• Multiple classification of resources under
  several classes
• Unordered, optional, and multi-valued properties
• Domain and range polymorphism of properties

51
A Cultural Community Resource Description Example
Portal Schema
Portal Resource Descriptions
r2 museoreinasofia.mcu.es/ guernica.jpg
r1www.rodin.fr/ thinker.gif
r4museoreinasofia.mcu.es
r3www.artchive.com/ woman.jpg
Web Resources
52
RDF/XML Serialization Data
ltrdfRDF xmllang"en" xmlnsrdf"http//w
ww.w3.org/1999/02/22-rdf-syntax-ns"
xmlnsrdfs"http//www.w3.org/TR/2000/PR-rdf-schem
a-20000327" xmlns""gt ltPainter
rdfidpicasso132"gt ltfnamegtPablolt/fnamegt
ltlnamegtPicassolt/lnamegt ltpaintsgt
ltPainting rdfabout"http//museoreinasofia.mcu.es
/guernica.gif"gt ltexhibitedgt
ltMuseum rdfabout"http//museoreinasofia.mcu.es"/
gt lt/exhibited gt
lttechniquegtoil on canvaslt/techniquegt
lt/Paintinggt lt/paintsgt ltpaintsgt
ltPainting rdfabout"http//www.artchive.com/woman
.jpg/gt lt/paintsgt lt/Paintergt ltExtResource
rdfabout"http//museoreinasofia.mcu.es"gt
lttitlegtReina Sophia Museumlt/title gt
ltlastmodifiedgt2000/06/09lt/lastmodifiedgt lt/ExtResou
rcegt ltSculptor rdfid"rodin424" lname"Rodingt
ltcreatesgt ltSculpture
rdfabout"http//www.rodin.fr/thinker.gif"/gt
lt/createsgt lt/Sculptorgt lt/rdfRDFgt
53
RDF/XML Serialization Schema
ltrdfRDF xmllang"en" xmlnsrdf"http//w
ww.w3.org/1999/02/ 22-rdf-syntax-ns"
xmlnsrdfs"http//www.w3.org/TR/2000/
PR-rdf-schema-20000327"gt ltrdfsClass
rdfID"Artist"/gt ltrdfsClass rdfID"Artifact"/gt
ltrdfsClass rdfID"Style"/gt ltrdfsClass
rdfIDMuseum"gt ltrdfsClass rdfID"Sculptor"gt
ltrdfssubClassOf rdfresource"Artist"/gt
lt/rdfsClassgt ltrdfsClass rdfID"Painter"gt
ltrdfssubClassOf rdfresource"Artist"/gt
lt/rdfsClassgt ltrdfsClass rdfID"Sculpture"gt
ltrdfssubClassOf rdfresource"Artifact"/gt
lt/rdfsClassgt ltrdfsClass rdfID"Painting"gt
ltrdfssubClassOf rdfresource"Artifact"/gt
lt/rdfsClassgt ltrdfProperty rdfID"creates"gt
ltrdfsdomain rdfresource"Artist"/gt
ltrdfsrange rdfresource"Artifact"/gt
lt/rdfPropertygt
ltrdfProperty rdfID"paints"gt ltrdfsdomain
rdfresource"Painter"/gt ltrdfsrange
rdfresource"Painting"/gt
ltrdfssubPropertyOf rdfresource"creates"/gt
lt/rdfPropertygt ltrdfProperty rdfID"sculpts"gt
ltrdfsdomain rdfresource"Sculptor"/gt
ltrdfsrange rdfresource"Sculpture"/gt
ltrdfssubPropertyOf rdfresource"creates"/gt
lt/rdfPropertygt ltrdfProperty rdfIDexhibited"gt
ltrdfsdomain rdfresource"Painting"/gt
ltrdfsrange rdfresourceMuseum"/gt lt/rdfProp
ertygt ltrdfProperty rdfID" technique"gt
ltrdfsdomain rdfresource"Painting"/gt
ltrdfsrange rdfresource"http//www.w3.org/
TR/1999/PR-rdf-schema-19990303Literal"/gt lt/rdfPr
opertygt ltrdfProperty rdfID"title"gt
ltrdfsdomain rdfresource"ExtResource"/gt
ltrdfsrange rdfresource "http//www.w3.org/
TR/1999/PR-rdf-schema-19990303Literal"/gt lt/rdfPr
opertygt . lt/rdfRDFgt
54
RDF/S vs. Well-Known Formalisms

• Relational or Object Database Models (ODMG, SQL)
• Classes dont define table or object types
• Instances may have associated quite different
  properties
• Collections with heterogeneous members
• Semistructured or XML Data Models (OEM, UnQL,
  YAT, XML Schema)
• Schema labels on both nodes and edges
• Class and property subsumption is not captured
• Heterogeneous structures reminiscent to SGML
  exceptions
• Knowledge Representation Languages (Telos, DL,
  F-Logic)
• Absence of complex values and n-ary
  relationships (bags, sequences)

55
A Formal Data Model for RDF

• ?(?,lt) is a well-defined hierarchy of
  classes/properties iff
• c ? C gt c lt Class
• p ? P gt p lt Property
• p1,p2 ? P and p1 lt p2 gt domain(p1) ? domain(p2)
  and range(p1) ? range(p2)
• Type System
• ? ?L ?U ? ? (1? 2? n?)
• Interpretation Function
• Literal types, ?L dom(?L)
• Bag types, ? ?1, ?2,, ?n, ?1, ?2,,
  ?n ? V are values of type ?
• Seq types, ? ?1, ?2,, ?n, ?1, ?2,,
  ?n ? V are values of type ?
• Alt types, (1?1 2?2 n?n ) ?I, ?i
  ? V, 1ltiltn is a value of type ?i
• c ? C, c ? ? ? ?(c)??(c) c lt c
• p ? P, p ?1, ?2 ?1 ? domain(p), ?1
  ? range(p)??(p) p lt p

56
A Formal Data Model for RDF

• An RDF schema is a 5-tuple RS (VS, ES, H, ?,
  ?)
• VS a set of nodes
• ES a set of edges
• ? (?,lt) a well-formed hierarchy of names
• ? an incidence function Es ? Vs?Vs
• ? a labeling function VS ? ES ?? ??
• An RDF description base, instance of a schema RS,
  is a 5-tuple RD (VD, ED, ?, ?, ?)
• VD a set of nodes
• ED a set of edges
• ? an incidence function ED ? VD?VD
• ? a valuation function VD ? V
• ? a labeling function VD ? ED ?2???
• ? u ? VD, ? ? n ? C?T ?(u) ?n
• ? e ? ED u,u, ? ? p

57
A Formal Data Model for RDF
H
Property
Class
lt
lt
?
N
?
?
?L
?
?C
T
?P
?
?
.
.
V
val,val
U
val
?
?
URI
?
S
resources
containers
literals
58
Some RDF Applications

• Web Browsers
• Netscape 6 from Netscape/AOL uses RDF to
  integrate various data-oriented applications such
  as bookmarks, mail/news, channels, etc. as well
  as for smart browsing and related links (RDF
  annotation services)
• Amaya Editor/Browser from W3C uses RDF to support
  user annotations on Web pages as metadata
• Brokers/Portals
• RSS (RDF Site Summary) XML/RDF Specification 1.0
  2000
• Web Service Description Language (WSDL) XML/RDF
  Specification 2000
• PICS Rating Vocabularies in XML/RDF W3C NOTE 27
  March 2000
• Platform for Privacy Preferences and RDF/RDF W3C
  Draft 10 May 2000
• Content Management
• OCLC Dublin Core Elements in RDF
• ICOM-CIDOC Conceptual Reference Model in RDF
• The Wordnet Lexical Ontology in RDF
• European Treasury Browser in RDF

59
Example Annotation Recommendation Services
60
Practical notes on RDF

• Authoring/Visualization
• by hand (experts only, perhaps copy paste)
• support by other tools (editors like Stanford
  Protégé)
• conversion from existing data stores (using XSLT)
• visualize RDF graphs (using Rudolf RDFViz)
• Parsing/Validating
• ICS-FORTH Validating RDF Parser (VRP)
• Rapier RDF Parser
• W3C Simple RDF Parser Compiler (SiRPAC)
• Storing/Querying
• ICS-FORTH RSSDB/RQL
• Aidministrator Sesame
• Redland Squish
• R.V.Guha RDFdb
• Harvesting/Crawling
• AIFB RDF Crawling

61
ICS-FORTH RDF RD Activities (http//www.ics.forth
.gr/proj/isst/RDF)
62
The ICS-FORTH RDFSuite

• The Validating RDF Parser (VRP) Karsten Tolle
  Diploma Thesis
• The first RDF Parser supporting semantic
  validation of both resource descriptions and
  schemas
• The RDF Schema Specific DataBase (RSSDB) Sophia
  Alexaki M.Sc. Thesis
• The first RDF Store using schema knowledge to
  automatically generate an Object-Relational
  (SQL3) representation of RDF metadata and load
  resource descriptions
• The RDF Query Language (RQL) Greg Karvournarakis
  M.Sc. Thesis
• The first Declarative Language for uniformly
  querying RDF schemas and resource descriptions

63
The ICS-FORTH RDFSuite Architecture
64
The Validating RDF Parser (VRP)

• The VRP parser checks only if an RDF file is
  well-formed according to the RDF M S Spec
• The VRP validator checks if the model (i.e.
  triples) generated by the parser satisfies the
  constrains imposed by the RDF Schema Spec

65
VRP Constraints (Schema)

• Class, Property and Type names are mutually
  exlusive
• C ? P ? T ?
• Literal, Resources and Container values are
  mutually exclusive
• L ? U ? V/U, L ?
• ? c, c? ? C
• Class is the root of class hierarchy
• c lt Class
• subClassOf relation is transitive
• c lt c?, c?lt c? ? ? c lt c? ?
• subClassOf relation is antisymmetric
• c lt c? ? c ? c?
• Domain and range of properties should be defined
  and they should be unique
• ?p ? P, ?!c1 ? C (c1 domain(p))
  ? ?!c2 ?C ? TL (c2range(p))

• ? p, p?, p?? ? P
• Property is the root of property hierarchy
• p lt Property
• subPropertyOf relation is transitive
• plt p?, p?lt p? ? ? plt p? ?
• subPropertyOf relation is antisymmetric
• p lt p? ? p ? p?
• If p is subPropertyOf of p? then domain of p is
  subset of domain of p? and range of p is subset
  of range of p?
• p lt p ?? domain(p) ? domain(p?) ?
  range(p) ? range(p?)
• A reified statement should have exactly one
  rdfpredicate, rdfsubject and rdfobject property

66
VRP Constraints (Data)

• For all values ?u ? V
• If u is a URI then it is an instance of one or
  more Classes
• u ? U ? ?(u) ? C
• If u is a literal then it an instance of one
  and only one Literal type
• u ? L ? ?(u) ? TL
• If u is a container then it an instance of one
  and only one Container type
• u ? V/U, L ? ?(u) ? TB S A

• For all properties ?p ? P, u1,u2 ? p
• if p belongs to the set 1, 2, 3 then u1 is an
  instance of either rdfBag or rdfSeq or rdfAlt
• if p ?1, 2, 3? ?(u1)?TB S A
• if p doesnt belong to 1, 2, 3, then u1
  belongs to the domain of p and u2 belongs to
  the range of p
• if p ? P/1, 2, 3 ? ?(u1) ? domain(p) ? ?(u2)
  ? range(p)

67
The RDF to DBMS Loader

• Resource
• URI

Extended VRP Validator
RDF Model

• RDF_Resource
• rdftype
• ...

P1
Persistent Namespace (DBMS)
Additional Constraints
C2
C1
store()
P1
r1
r2
RDF Querying APIs

• RDF_Statement
• rdfpredicate
• rdfsubject
• rdfobject

• RDF_Property
• rdfsdomain
• rdfsrange
• rdfssubPropertyOf
• link_list

• RDF_Class
• rdfssubClassOf

store()
store()
RDF Loading APIs
DBMS
r1
68
RSSDB Representation of RDF metadata
Class
Property
id
nsid
lpart
nsid
lpart
domainid
rangeid
id
14
4
title
10
1
10
DataResource
15
11
2
ExternalPage
2
title
11
1
3
12
Arts
SubClass
3
13
Art_History
SubProperty
subid
superid
11
10
subid
superid
12
10
15
14
13
12
69
The RDF Query Language (RQL)

• Declarative query language for RDF description
  bases
• relies on a typed data model (literal container
  types union types)
• follows a functional approach (basic queries and
  filters)
• adapts the functionality of semistructured or XML
  query languages to RDF, but also
• treats properties as self-existent individuals
• exploits taxonomies of node and edge labels
• allows querying of schemas as semistructured data
• Relational interpretation of schemas resource
  descriptions
• Classes (unary relations)
• Properties (binary relations)
• Containers (n-ary relations)

70
Browsing Portal Catalogs with RQL

• Simple set queries on class and property extents
• Find the resources in the extent of the property
  creates
• creates
• www.portal.gr/rodin424, www.rodin.fr/thinker.
  gif,
• www.portal.gr/picasso132,
  museoreinasofia.mcu.es/guernica.gif,
• www.portal.gr/picasso132,
  www.artchive.com/woman.jpg
• Find the resources of type painter and sculptor
• ExtResource intersect Sculpture
• www.rodin.fr/thinker.gif
• Schema constructs used as query terms support
  for automatic query expansion (similar to
  thesauri-based IRS)
• Useful to query resources with minimal schema
  knowledge

Includes paints sculpts
Multiply classified resources
71
Personalizing Portal Catalogs with RQL

• Navigational queries on semistructured resource
  descriptions
• Find the Museum resources that have been modified
  in year 2000.
• select x
• from Museumx.last_modifiedy
• where y gt 2000/01/01
• museoreinasofia.mcu.es
• Similar functionality to semistructured or XML
  query languages (Lorel, UnQL, XQL, XML-QL,
  XML-GL)
• Useful in the absence of schema information or
  when multiple schemas are used to describe
  resources

Data paths not foreseen in the schema
72
Querying Portal Catalogs with Large Schemas

• Filtering both resource descriptions and schemas
• Find the paintings having as technique oil on
  canvas that have been created by a
  neo-impressionist painter
• select y
• from XcreatesyPainting.techniquez
• where X lt neo-impressionist and z oil on
  canvas

Data filtering with schema information
Schema Filtering on Class hierarchies
73
Querying Portal Schemas with RQL

• Pure schema queries
• Find the properties which specialize the property
  creates and may have as domain the class Painter
  along with their corresponding range classes
• select _at_P, Y
• from Painter_at_PY
• where _at_P lt creates
• creates, Artifact,
• creates, Painting,
• creates, Sculpture,
• paints, Painting

All Properties defined or inherited in class
Painter
Schema filtering on property hierarchies
74
RQL Examples

• Similar functionality to DBMS schema QLs
  (SchemaSQL, XSQL)
• Useful for large schemas (integrating ontologies
  and thesauri)

75
Putting it all Together

• Nested schema and data queries
• Find the resources modified after 2000/01/01
  which can be reached by a property applied to the
  class Painting and its subclasses
• select R, y
• from (select _at_P
• from X_at_P
• where X lt Painting)R.ylast_modif
  iedz
• where z gt 2000/01/01
• exhibited, museoreinasofia.mcu.es
• Subcommunities may use different schemas while
  sharing the same description base

R ranges over the labels of type property
76
RQLExamples
Portal Schema
Portal Resource Descriptions
77
Putting it all Together

• Schema and data queries
• Find all metadata about the resources of the site
  museoreinasofia.mcu.es
• select x,Y,P,z,W
• from xYPzW
• where x like museoreinasofia.mcu.es or
• y like museoreinasofia.mcu.es
• www.portal.gr/picasso132, Painter, paints,
  museoreinasofia.mcu.es/guernica.gif, Painting,
• museoreinasofia.mcu.es/guernica.gif,
  Painting, exhibited, museoreinasofia.mcu.es,
  Museum,
• museoreinasofia.mcu.es/guernica.gif,
  Painting, technique, oil on canvas, string,
• museoreinasofia.mcu.es, ExtResource, title,
  Reina Sophia Museum, string,
• museoreinasofia.mcu.es, ExtResource,
  last_modified, 2000/06/09, date, .
• Subcommunities may use both different schemas and
  description bases

URLs pattern matching
78
RQL Query Processing
select y from xcreatesyPainting.has_material
z where z oil on canvas
select y from creates A, has_material B, D C
define x A.source, y A.target, w B.source,
z B.target, R range(creates), D
subclassOf(R), E (C) where z oil on
canvas and y w and C Painting and y in E
79
RQL Query Optimization
select X.target from creates X, has_material Y,
Painting P where X.target Y.source and X.target
P.uri and Y.target oil on canvas
80
The RQL Query Interpreter
Syntax analysis
Query string

• Syntactical analysis (lex/yacc)
• CNF transformation

(1)
Type inference
Syntax tree under CNF

• Checks type compatibility
• Sets appropriate evaluation functions

(2)
Query string
(3)
Main
Graph construction
Query result
(4)

• Evaluation of dependencies
• Factorization functions

Query graph
Typing
DBMS RDF Query APIs
(5)
Query graph
Evaluator
Evaluation
(6)

• Defines evaluation functions
• Query Processing

Result
DBMS
81
RDFSuite Summary

• RDFSuite addresses the needs of effective RDF
  metadata management by providing tools for
  validation, storage and querying
• validation follows a formal data model and
  constraints enforcing consistency of RDF schemas
• incremental loading of voluminous description
  bases in a persistent store
• declarative query language for schema and data
  querying
• Ongoing efforts
• RQL query optimization
• transactional aspects
• alternative encoding and representation schemes
  for access optimization

82
Acknowledgements

• Funding was generously provided by the projects
• C-WEB (IST-1999-13479) A Generic Platform
  Supporting Community Webs
• MESMUSES (IST-2000-26074) Metaphor for Science
  Museums

83
Thank you
Hvala
Danke
Merci
Gracias
Grazie