Semantic Web

1
Semantic Web

• Presented by
• Edward Cheng
• Wayne Choi
• Tony Deng
• Peter Kuc-Pittet
• Anita Yong

2
Agenda

• Introduction
• Semantic Web components
• Example
• Pros/Cons

3
The Current World Wide Web

• Network of hypertext documents containing text,
  images, audio, and video
• Is readable and understandable by humans, but
  only readable (via parsing) by machines
• What is missing is this understanding of the
  semantics
• For example

vs.
ltIMGgt
4
Semantics

• The definition of semantics is meaning
• The idea is to give a keyword (or a subject) a
  definition
• These definitions are then related to objects
• Humans do this easily but often make mistakes, so
  to have computers perform these operations would
  eliminate mistakes
• Does not mean that there are no mistakes in the
  algorithms used, but idea is that using the logic
  of Semantic Web these errors could be caught

5
What is Semantic Web?

• Tim Berners-Lees vision of a network of data
  fully understandable by machines without
  requiring human intervention
• Standardized way of defining relations between
  things that have meaning
• A globally and semantically linked network of
  information underlying the current Web

6
Wedding Cake
7
Components

• Unicode
• A standard and universal way of encoding text
• URI (Uniform Resource Identifier)
• A unique string identifying a name or resource
• XML (Extensible Markup Language)
• Used to describe and communicate data

8
Components

• Namespaces
• A way of encapsulating a vocabulary
• XML Query
• Used to extract data from an XML document
• XML Schema
• A description of an XML document, imposing
  constraints on its content and structure

9
Components

• RDF Model (Resource Description Framework)
• Makes a statement about a resource using a
  (subject, predicate, object) URI triple
• (http//santa.com, http//xmlns.com/foaf/0.1/term
  _interest, http//en.wikipedia.org/wiki/Dieting
• Santa is interested in Wikipedias dieting entry

10
Ontology

• An ontology is a data model that describes
  individuals, classes, attributes, and relations
• RDF schema is the base RDF ontology, putting
  constraints on an RDF subjects relationship to
  an object for a given predicate
• foafperson foafknows foafperson
• foafimage foafknows foafperson
• OWL (Web Ontology Language)
• A more complicated and expressive ontology than
  RDF Schema
• Three forms OWL Lite, OWL DL, OWL Full

11
Querying

• A Semantic Web query connects to multiple RDF
  dataset agents to gather information, and returns
  the set of best matched results.

Cheapest Support Labour?
Agent
Agent
Agent
North American Firm A
North American Firm B
Overseas Firm A
12
Bob Book doctors appointment
Target Person
Type Patient
Type Doctor
Target Schedule
Type Bob
Type Rick
Type Sched
Type Sched
Time
Time
Time
Time
Employee
Target Hospital
13
Pros

• Searching with semantic meaning and relations
• Agents (programs) acting as intermediaries to
  meet user needs
• Better knowledge management as data is connected
  in meaningful ways

14
Cons

• Need to maintain two different versions of web
  content a machine readable version and a human
  interaction version
• Only useful if Semantic Web is adopted by the
  masses

15
Questions?