The semantic web

1
The semantic web I. Introduction What is
the semantic web II. Components of the
semantic web Markup and metadata
languages Ontologies and agents
III. What is it good for?
Agent-based transactions
Business applications Digital
libraries Personal information
management
2
The semantic web I. What is the semantic
web? The problem Most of the content on the web
today is designed for humans to read It is not
designed for computer programs to manipulate
meaningfully Computers can adeptly parse Web
pages for layout and routine processing This
is a header, a link to another page In general,
computers have no reliable way to process the
semantics of web pages
3
The semantic web is the abstract representation
of data on the World Wide Web, based on the RDF
standards and other standards to be defined It
is being developed by the W3C, in collaboration
with a large number of researchers and
industrial partners. W3C. (2001). Semantic Web
Introduction. http//www.w3.org/2001/sw The
semantic web is an extension of the current web
in which information is given well-defined
meaning, better enabling computers and people to
work in cooperation. Berners-Lee, T., Hendler,
J., and Lassila, O. (2001). The Semantic Web.
Scientific American
4
The web can reach its full potential only if it
becomes a place where data can be shared and
processed by automated tools as well as by
people For the web to scale, tomorrows programs
must be able to share and process data even when
these programs have been designed totally
independently The semantic web is a vision The
idea of having data on the web defined and linked
in a way that it can be used by machines not
just for display purposes, but for automation,
integration and reuse of data across various
applications W3C. (2001). Semantic Web
Introduction. http//www.w3.org/2001/sw
5
The semantic web will be an extension of the
current web It will bring structure to the
meaningful content of Web pages It will create
an environment where software agents roaming
from page to page can readily carry out
sophisticated tasks for users Machines will
become much better able to process and
understand the data that they currently can
only display It will operate alongside the
traditional web
6
The semantic web will be a web of trust All
statements found on the Web occur in some
context Applications need this context in order
to determine the trustworthiness of the
statements The machinery of the semantic web,
does not assert that all statements found on
the Web are true Truth - or more
pragmatically, trustworthiness - is evaluated
by, and in the context of, each application
that processes the information found on the Web
7
The semantic web I. Introduction What is
the semantic web II. Components of the
semantic web Markup and metadata
languages Ontologies and agents
III. What is it good for?
Agent-based transactions
Business applications Digital
libraries Personal information
management
8
II. Components of the semantic web For the
semantic web to function, computers must have
access to structured collections of information
and sets of inference rules The challenge is to
provide a language that expresses data and rules
for reasoning about the data It requires a
decentralized system of knowledge
representation It should allow rules from any
existing knowledge representation system to be
exported to the web This means adding logic
to the web Computers will use rules to make
inferences, choose courses of action and answer
questions
9
Four important technologies for developing the
semantic web are already in place XML It allows
users to add arbitrary structure to documents
but says nothing about what the structures
mean Resource Description Framework (RDF) RDF
provides the meaning Web ontologies They prevent
collisions and clarifies meaning Agents These
are programs that collect web content from
diverse sources
10
Markup and metalanguages The semantic web uses
uniform resource identifiers (URI) A simple and
extensible means for identifying a resource URL
a subset identifying resources by their primary
access mechanism (network location) Uniform
Resource Name (URN) a subset required to
remain globally unique and persistent even when
the resource ceases to exist or becomes
unavailable With URIs, everything can be on the
Web People, places, and things in the physical
world will have online representations
identified by URI This will facilitate effective
integration, active participation and be
contextualized in the semantic web
11
Common URIs http// Scheme for Hypertext
Transfer Protocol services mailto Scheme for
electronic mail addresses news Scheme for
USENET news groups and articles telnet// Scheme
for interactive services via the TELNET
Protocol ftp// Scheme for FTP
services gopher// Scheme for Gopher and Gopher
Protocol services The W3C maintains an extensive
listing of URI schemes (86) http//www.w3.org
/Addressing/schemes
12
Resource Description Framework (RDF) A framework
for describing and interchanging
metadata Resources things that can be named
with URIs Description statements about the
properties and values of these
resources Framework a common model for
diversity RDF aims to build a web of overlapping
metadata vocabularies It uses URIs and XML to
define metadata vocabularies It is a self
describing framework for describing
resources Published as the Resource Description
Framework (RDF) Model and Syntax Specification
W3C Recommendation, 22 February 1999
13
How RDF is designed Independence Since a
property is a resource, anyone can invent
them Scalability RDF statements are simple,
three-part records so they are easy to handle
and look things up by, even in large numbers
This is a triple resource, property,
value Interchange RDF statements can be
converted into XML and are easy to interchange
14
Properties are resources Properties can have
their own properties and can be found and
manipulated like any other resource Values can be
resources Values of properties, which include
title and authors name, have to include
resources Most web pages will have a property
named home page which points to a page on
their site Statements can be resources Statements
can have properties A statement is a triple
containing of a subject, a property and an object
15
Statement Subject Home page We need a way to
identify the thing we want to describe Property
(or predicate) We need a way to
identify a specific property of the thing Object
Best damned home page on the web We need a way
to identify the thing we want to assign as the
value of this property As RDF, it would look
like this triple title,
http//www.slis.indiana.edu/hrosenba/Home/home.ht
ml, Best damned home page on the
web
16
What it looks like xmlnsrdf"http//www.w3.org/1999/02/22-rdf-synta
x-ns xmlnsex"http//www.exam
ple.org/terms/" rdflocation" http//www.slis.indiana.edu
/hrosenba/Home/home.html " Best
damned home page on the web
This
example calls the RDF namespace It uses the
subject, property, and value from the triple
17
RDF provides a common framework for representing
metadata across many applications It is a
meta-language that provides a model and syntax
for metadata It is a means for developing tools
and applications using a common syntax for
describing web resources. This allows
independent parties to exchange and use it It
doesnt provide any properties of its own The
semantics and structure of many types of metadata
will be specified by independent communities It
provides an infrastructure that is sufficiently
general and flexible to support these disparate
applications
18
RDF metadata can be used in a variety of
application areas Resource discovery To provide
better search engine capabilities Cataloging For
describing the content and content relationships
available at a particular web site, page, or
digital library Intelligent software agents To
facilitate knowledge sharing and exchange Content
rating To label collections of pages IP To
describe digital rights management of web pages
19
Uses Trust RDF with digital signatures will be
key to the web of trust for ecommerce,
collaboration, and other apps Search and
retrieval A uniform query capability and
language for retrieving metadata from third
parties Processing rules language for automated
decision- making about web resources RDF
provides the basis for generic tools for
authoring, manipulating, and searching machine
understandable data on the web The goal is to
transform the web into a machine-processable
repository of information
20
How a computer sees the web without RDF
http//www.w3.org/Talks/2001/12-semweb-offices/sli
de8-0.html
21
How a computer sees the web with RDF
http//www.w3.org/Talks/2001/12-semweb-offices/sli
de9-0.html
22
Ontologies and agents An ontology is a file
that formally defines the relations among
terms The most typical kind of web ontology has
a taxonomy and a set of inference rules An
agent comparing or combining information across
multiple databases has to know that synonymous
terms mean the same thing Also, it has to know
that the same term can have different meaning
in different databases Ontologies provide a way
to discover such common meanings for whatever
databases the agent encounters
23
A common ontology defines the vocabulary used in
the queries and assertions that are exchanged
among agents An ontological commitment is an
agreement to use the shared vocabulary in a
coherent and consistent manner Agents sharing a
vocabulary need not share a knowledge base
Each knows things the other does not An agent
committing to an ontology is not required to
answer all queries that can be formulated in
the shared vocabulary A commitment to a common
ontology is a guarantee of consistency, but not
completeness Gruber, T. (nd). What is an
Ontology? http//www-ksl.stanford.edu/kst/what-
is-an-ontology.html
24
A taxonomy defines classes of objects and
relations among them A taxonomy expresses a
large number of relations among entities
This occurs because it assigns properties to
classes and allows subclasses to inherit
these properties The meaning of terms or XML tags
used on a web page can then be defined by
pointers from the page to an ontology A
collision can occur when gathering data from two
sites with two ontologies that use different
terms having the same meaning This can be
resolved if the ontologies provide equivalence
relations that equate the two terms
25
Inference rules allow computers to deduce data
and make decision Consider the rule If a city
code is associated with a state code, and an
address uses that city code, then that address
has the associated state code Given an IU
address, an agent could then deduce The IU
address has Bloomingtons city code It must be
in Indiana, which is in the U.S. It should be
formatted to U.S. standards
26
Ontologies can enhance the functioning of the web
They can be used in to improve the accuracy of
web search and retrieval The search program
looks for only those pages referring to a
precise concept instead of the ones using
ambiguous keywords More advanced applications
will use ontologies to relate the information on
a page to associated knowledge structures and
inference rules I could have an ontology that
defines the various classes I teach and the
relations among them An agent could then easily
find related syllabi on my site from my home page
27
Examples of web ontologies OWL (Web Ontology
Language) A semantic markup language for
publishing and sharing ontologies and their
associated knowledge bases on the web Uses RDF
and DAMLOIL Used to describe the classes and
relations between them inherent in web documents
and applications OWL includes Taxonomic
relations between classes Datatype properties,
descriptions of attributes of elements of
classes
28
OWL also includes Object properties Descriptions
of relations among elements of
classes Instances of classes Instances of
properties Datatype and object properties are
the properties of a class A set of OWL
assertions loaded into a reasoning system is
called a knowledge base These may include facts
about individuals members of classes, derived
facts, and facts not literally present the
ontology, but logically implied by the semantics
of OWL http//www.w3.org/TR/2002/WD-owl-ref-20021
112/
29
DAMLOIL (DARPA Agent Markup LanguageOntology
Interchange Language) A semantic markup language
for web resources using RDF and RDF Schema It
consists of zero or more headers, followed by
zero or more class elements, property elements,
and instances The header contains version and
importing tags, which reference additional
ontologies There is a datatype domain containing
XML schema There is an object domain containing
classes and subclasses of the DAMLOIL
ontology http//www.w3.org/TR/2001/NOTE-damloil-
reference-20011218
30
SHOE (Simple Ontology HTML Extensions) It is an
HTML-based knowledge representation language that
adds the tags needed to embed arbitrary semantic
data into web pages SHOE tags are divided into
two categories There are tags for constructing
ontologies These rules define the kinds of
assertions SHOE documents can make and what
these assertions mean There are tags that
Annotate documents to subscribe to other
ontologies Declare data entities Make
assertions about entities under the rules of the
ontologies
31
SHOE is primarily meant to make it possible for
web robots and intelligent agents It is a
meta-language for ontologies Particular
categories, relationships, attributes,
inferences, etc. are not defined by SHOE SHOE
ontologies declare Classifications (categories)
for data entities Classifications may inherit
from other classifications (Dogs are
Animals) Valid relationships between data
entities and other data entities or simple data
(strings, numbers, dates, booleans) http//www.
cs.umd.edu/projects/plus/SHOE/
32
Classifications Arguments for valid
relationships use simple data or the
classification a data entity falls under to fill
an argument ("Dogs can chase cats"). Inferences
in the form of clauses with no negation ("If a
person works for an organization, that person
works for any organization the organization is a
sub-part of"). Inheritance from other
ontologies Ontologies may be derived from or
extend zero or more outside ontologies ("The
SPCA ontology extends the common Library of
Congress ontology"). Versioning Ontologies may
extend previous ontology versions
33
HTML pages with embedded SHOE data may Declare
arbitrary data entities Usually, one of these
entities is the web page itself Declare the
ontologies they will use when making
declarations about entities (Im using the
ASPCAs Pets ontology). Categorize entities
(This entity is a dog) Declare relationships
between entities or between entities and data
(This entity is Fido) Agents using SHOE are
written to assume that declarations made by
entities are claims and not simple facts
34
Architecture of the semantic web
http//www.w3.org/2001/09/06-ecdl/slide17-0.html
35
The semantic web I. Introduction What is
the semantic web II. Components of the
semantic web Markup and metadata
languages Ontologies and agents
III. What is it good for?
Agent-based transactions
Business applications Digital
libraries Personal information
management
36
III. What is it good for? The semantic web will
allow us to realize the full potential of the
Web Everything will be identifiable on the
semantic web It will make it cost-effective for
people to effectively record their
knowledge The ultimate goal of the semantic
web The design of enabling technologies to
support machine facilitated global knowledge
exchange Miller, E. (2000). Goals of the
semantic web. http//www.w3.org/2001/09/06-ecdl/sl
ide8-0.html
37
Semantic web
38
Agent-based transactions There will be
customizable programs that will be able to
Collect Web content from diverse
sources Process the information Exchange the
results with other programs The value of agents
is subject to network externalities Agents will
be able to interact if they are written to
process semantic content of web data They take
advantage of the trust built into the semantic
web Agents inference engines can exchange and
recognize proofs based on inferences and rules
39
Agents for individuals and web services will be
able to interact by coming to a common
understanding What one wants - what the other
has This requires a common language Agents
will exchange ontologies providing each other
with the vocabulary needed for discussion A
semantic value chain can be established among
linked agents that will assemble the information
requested by the individuals agent
40
Agents will be able to engage in service
discovery The semantic webs common language
will be used to describe a service This will
allow agents understand the functions offered
by the services and how to take advantage of
them Agents operate in a web of trust All
information on the Web occurs in a context
Agents need access to this context to evaluate
the trustworthiness of the information Trustwort
hiness is evaluated by each application that
processes the information
41
This will involve the exchange of "proofs"
written in the semantic webs unifying
language This is the language using RDF, OWL,
and XML that expresses logical inferences made
using rules and information such as those
specified by ontologies An agent requesting a
proof is asking for the chain of reasoning
provided by the other agents inference
engine This internal reasoning is checked by
the first agent and accepted or rejected Some
programs can already exchange proofs in this way,
using the current preliminary versions of the
unifying language
42
When this is formalized, agents will be able to
access trusted services This will allow
automated checking an verification of service
and resource locations on the net Your agent
makes you an airline and hotel reservation
after checking to make sure that the businesses
meet your standards This will use digital
signatures These will be used to verify that the
attached information has been provided by a
specific trusted source Agents will be written
to be skeptical of assertions that read on the
semantic web until they have checked the
sources of information
43
Business applications The semantic web will
provide a layer of interoperability that will
allow systems to transparently exchange
data DigitalOwls KineticEdge is a content
management and publishing solution http//www.
digitalowl.com Epigraphs Xcellerant is a content
management and ontology management
solution http//www.epigraph.com


Ontoprises Ontobroker is inference
middleware http//www.ontoprise.de Unicorn
Solutions Unicorn Coherence is ontology modeling
and data integration software http//www.unicorn
.com
44
Digital libraries There is a convergence of
digital libraries, museums and archives as memory
organizations There are technical and social
areas of overlap between semantic web and
digital libraries It is critical to help these
memory organizations develop as key foundations
for supporting the semantic web For example,
there is a Semantic Web Advanced Development
collaboration with the DSpace project Dspace is
a joint project of MIT Libraries and HP Its
goal is to develop a flexible digital archive
designed capture and distribute the intellectual
output of MIT
45
Personal information management This is a
specialized version of agent based
transactions It takes full advantage of
convergence and networked information
appliances Our agents will be able to use
characteristics of the semantic web to manage
many of our routine tasks The devices in our
home use have descriptions of services and
functionalities Our agents can understand
these and make use of them The phone rings
and the TV volume is turned down when you
pick up
46
Collaborative research Research that is
computationally intensive will be able to be
handled easily on the semantic web Machine to
machine processing and exchange of data is one
of the basic functionalities