Representing%20Data%20with%20XML

1
Representing Data with XML

• September 27, 2005
• Shawn Henry
• with slides from Neal Arthorne

2
Data Representation

• Design goals for data representation
• Portable (platform independent)
• Easy for machines to process
• Human legible
• Flexible and usable over the Internet and other
  networks
• Concisely defined with formal rules

3
Extensible Markup Language

• World Wide Web Consortium (W3C) defines the
  Extensible Markup Language (XML)
• W3C also defined HTML, CSS, HTTP, SVG and other
  markup languages
• XML Working group formed in 1996
• XML 1.0 (Third Edition) 4 February 2004 (original
  Recommendation in 1998)

4
XML Example

• lt?xml version"1.0" encoding"UTF-8"?gt
• ltfoodsgt
• ltpizza titleDeluxe Pizzagt
• ltnamegtThe Deluxelt/namegt
• lttoppingsgt
• lttoppinggtpepperslt/toppinggt
• lttoppinggtpepperonilt/toppinggt
• lttoppinggtmushroomslt/toppinggt
• lttoppinggtcheeselt/toppinggt
• lttoppinggttomato saucelt/toppinggt
• lt/toppingsgt
• ltpricegt7.99lt/pricegt
• lt/pizzagt
• lt/foodsgt

5
XML

• XML documents should be well-formed (syntax,
  closing tags etc)
• XML documents are valid if they conform to a
  specified grammar (usually DTD or XML Schema)
• DTDs (Document Type Definitions) provide a
  grammar for the XML by defining elements,
  attributes and entities

6
XML Advantages

• XML provides
• Logical structure for data in a textual
  representation
• Formal rules for validating documents
• Flexibility to define your own markup language
• Portability across networks and platforms
• Becoming a widely accepted data interchange
  format
• Processed with off-the-shelf tools

7
XML Disadvantages

• XML drawbacks
• Not a binary format so it requires a lot of
  overhead for a little bit of data
• Very little support for binary or mixed media
  data formats (hex or base64 encoding)
• Only for data and holds no semantics or reasoning
• DTDs do not provide
• Data types for each element or attribute
• Complex structural rules for documents

8
XML Schema

• XML Schema defines a new schema language to
  replace DTD
• Standardized by W3C in 2001
• Advantages
• Provides data typing and logical structure
• Written in XML (easy to process)
• Higher complexity than DTD

9
XML Schema Example

• lt?xml version"1.0" encoding"UTF-8"?gt
• ltxsdschema xmlnsxsd"http//www.w3.org/2001/XMLS
  chema"gt
• ltxsdelement name"pizza"gt
• ltxsdcomplexTypegt
• ltxsdallgt
• ltxsdelement name"name" type"xsdstring" /gt
• ltxsdelement name"toppings" type"Toppings"
  /gt
• ltxsdelement name"price" type"xsdfloat" /gt
• lt/xsdallgt
• ltxsdattribute name"title" type"xsdstring"
  /gt
• lt/xsdcomplexTypegt
• lt/xsdelementgt
• ltxsdcomplexType name"Toppings"gt
• ltxsdsequencegt
• ltxsdelement name"topping" minOccurs"1"
  maxOccurs"unbounded" type"xsdstring" /gt
• lt/xsdsequencegt
• lt/xsdcomplexTypegt
• lt/xsdschemagt

• An XML document is an instance document of an
  XML Schema

10
Simple Types

• Simple Types are of three varieties
• Atomic Built-in or derived, e.g.
• ltxsdsimpleType name"myInteger"gt
• ltxsdrestriction base"xsdinteger"gt
• ltxsdminInclusive value"10000"/gt
• ltxsdmaxInclusive value"99999"/gt
• lt/xsdrestrictiongt
• lt/xsdsimpleTypegt
• List multiple items of the same type
• ltlistOfMyIntgt20003 15037 95977 95945lt/listOfMyIntgt
  
• Union Union or two or more Simple Types

11
Built-in Types

• XML Schema defines numerous built-in types
• integer, decimal, token, byte, boolean, date,
  time, short, long, float, anyURI, language
• Facets can be used to restrict existing types
• min/maxInclusive, min/maxExclusive, pattern,
  enumeration, min/maxLength, length, totalDigits,
  fractionDigits

12
Complex Types

• Complex Types define logical structures with
  attributes and nested elements
• They use a sequence, choice or all containing
  elements that use Simple Types or other Complex
  Types
• May reference types defined elsewhere in the
  schema or imported using import statement

13
In the Schema of Things

• XML Schema supersedes DTD
• Defines a typed data format with no semantics or
  relations between data
• Next step higher level of abstraction and the
  ability to define objects and relations

14
Resource Description Framework

• W3C standard for describing resources on the
  World Wide Web (1999, revised 2004)
• Objects identified by Uniform Resource
  Identifiers (URIs)
• Generalized to identify objects that may not be
  retrievable on the Web
• RDF represented by a directed graph and in XML
  syntax

15
RDF Example
http//www.example.com/people/diaz/contact
http//www.w3.org/2000/10/pim/contactfullName
http//www.w3.org/2000/10/workemployer
Federico Diaz
http//www.fisherandsons.com/contact

• In English http//www.example.com/people/diaz/con
  tact has the full name Federico Diaz and has an
  employer called Fisher and Sons.

16
RDF Parts

• Each RDF statement is a triple containing a
  subject (identifier by URI), a predicate (e.g.
  creator, title, full name) and an object
• An object can be either a literal value (e.g.
  Federico Diaz) or another RDF resource
• All three parts can be identified with an URI and
  fragment identifier

17
RDF Semantics

• RDF attaches no specific meaning to RDF
  statements just like the name of a database
  field is meaningless to an SQL engine
• RDF does provide a way to attach data types to
  literal values, but RDF does not define data
  types
• Generally RDF software uses the XML Schema data
  types
• ltsize rdfdatatypexsdintgt10lt/sizegt
• Arbitrary XML can also be used as a literal
• ltxprop rdfparseType"Literalgt
  ltasizegt10lt/asizegtlt/xpropgt

18
RDF Schema

• RDF Schema is a vocabulary description language
  that relates resources to each other using RDF
• RDFS uses classes of objects like in
  Object-Oriented (OO) systems
• Class properties relate to other classes using OO
  concepts such as generalization

19
RDF Schema Use

• Differs from OO in that Properties are defined in
  terms of the resources to which they apply (their
  domain) they are not restricted to the scope of
  a single class
• domain Classes to which a Property applies
• range The Class of a Property (i.e. type)
• Allows new Properties to be created that apply to
  the same domain without redefining the domain

20
RDFS Classes

• Classes introduced by RDFS
• Resource - top level class
• Literal all literal values like text strings
• Class the class of all classes
• Datatype top level RDF datatype
• Properties introduced by RDFS
• subClassOf
• subPropertyOf
• domain domain of a Property
• range range of a Property
• label, comment, seeAlso human readable labels

inheritance
21
RDFS Example

• lt?xml version"1.0"?gt
• lt!DOCTYPE rdfRDF lt!ENTITY xsd
  "http//www.w3.or/2001/XMLSchema"gtgt
• ltrdfRDF
• xmlnsrdf"http//www.w3.org/1999/02/22-rdf-synt
  ax-ns"
• xmlnsrdfs"http//www.w3.org/2000/01/rdf-schema
  "
• xmlbase"http//example.org/schemas/food"gt
• ltrdfsClass rdfID"Food"/gt
• ltrdfsClass rdfID"Pizza"gt
• ltrdfssubClassOf rdfresource"Food"/gt
• lt/rdfsClassgt
• ltrdfsClass rdfID"Topping"gt
• ltrdfssubClassOf rdfresource"Food"/gt
• lt/rdfsClassgt
• ltrdfsDatatype rdfabout"xsdfloat"/gt
• ltrdfProperty rdfID"hasTopping"gt

22
RDF Example

• lt?xml version"1.0"?gt
• lt!DOCTYPE rdfRDF lt!ENTITY xsd
  "http//www.w3.org/2001/XMLSchema"gtgt
• ltrdfRDF xmlnsrdf"http//www.w3.org/1999/02/22-r
  df-syntax-ns"
• xmlnsex"http//example.org/schemas/f
  ood"
• xmlbase"http//example.org/things"gt
• ltexPizza rdfID"ShawnsPizza"gt
• ltexprice
• rdfdatatype"xsdfloat"gt12.99lt/expri
  cegt
• ltexhasTopping rdfresource"http//www.exa
  mple.org/food/85740"/gt
• ltexhasTopping rdfresource"http//www.exampl
  e.org/food/85729"/gt
• lt/exPizzagt
• lt/rdfRDFgt

23
RDF/RDFS

• Lets authors create vocabularies of Classes and
  Properties and show how the terms should be used
  to describe resources, e.g.
• Property author applies to class Book
• Class Employee is a subclass of Person
• Does not define descriptive properties such as
  dateOfIssue or title but references them
  using URIs
• Like in XML/XML Schema, an RDF instance document
  can be validated against its RDF Schema

24
Machines Understanding the Web

• RDF/RDFS along with XML/XML Schema provide a
  means to describe resources on the web with basic
  generalization
• For a higher conceptual level, applications
  require semantic information
• Ontologies serve as a starting point for
  understanding

25
Ontologies on the Web

• Ontologies define the terms used to represent an
  area of knowledge. OWL Use Cases
  Requirements, 2004
• Example use cases
• A web portal that needs to classify information
• Multimedia archive that requires a taxonomy of
  media or content-specific properties
• Corporate portal website that integrates
  vocabularies from different departments

26
Web Ontology Language (OWL)

• Supersedes DAMLOIL
• DARPA Agent Markup Language (DAML) was based on
  RDF/RDFS and includes much of what is now OWL
• Adds terms used to better describe relations
  between classes of RDF resources
• With OWL, ontologies can be integrated, extended
  and shared

27
Web Ontology Language

• Individuals
• OWL does not honour the Unique Names Assumption
  (UNA)
• Properties
• Binary relations between individuals
• Functional, transitive or symmetric
• Classes
• Sets containing individuals
• Organized into a taxonomy with subclasses and
  superclasses

28
Three Flavours of OWL

• OWL Lite
• For classification hierarchies with simple
  constraints
• OWL DL
• Expressiveness with computational completeness
• OWL Full
• Maximum expressiveness
• No computational guarantees
• Extension of RDF

29
OWL Features

• OWL improvements on RDF/RDFS
• Cardinality
• min/maxCardinality for Properties with respect to
  a Class
• Equality, disjointness
• equivalentClass, equivalentProperty, sameAs,
  differentFrom, disjointWith
• Transitive, Symmetric, Functional Properties
• labelling a Property allows for reasoning
• A has B and B has C implies A has C (Transitive)
• A has B implies B has A (Symmetric)

30
OWL Features (contd)

• Boolean expressions of Class relations
• unionOf, complementOf, intersectionOf
• Property restrictions
• Limits how properties can be used by an instance
  of a class
• Versioning
• priorVersion, versionInfo, incompatibleWith,
  backwardCompatibleWith

31
Conclusion
???
Conceptual level reasoning smart applications
OWL
Knowledge processing and reasoning
RDF
RDF Schema
Resource description and vocabulary
Knowledge Data
XML
XML Schema
Data formatting and data types
Unicode/ISO byte streams
Machine data representation
32
References

• World Wide Web Consortium http//www.w3.org
• XML http//www.w3.org/TR/REC-xml
• XML Schema Part 0 Primer http//www.w3.org/TR/xml
  schema-0/
• RDF Primer http//www.w3.org/TR/rdf-primer/
• RDF Concepts http//www.w3.org/TR/rdf-concepts/
• RDF/XML Syntax http//www.w3.org/TR/rdf-syntax-gra
  mmar/
• RDF Schema http//www.w3.org/TR/rdf-schema/
• OWL Use Cases Requirements http//www.w3.org/TR/
  webont-req/
• OWL Overview http//www.w3.org/TR/owl-features/