Semantic Web

wrendeceitInternet and Web Development

Oct 21, 2013 (3 years and 5 months ago)

49 views

1

Semantic Web

Jayasree, Jenny, Raj, Sunil

2

Agenda


What is Semantic Web vs Semantic


History of Semantic Web


Current Research Areas (eg OWL, RDF,
reasoning engines, query languages)


Applications (eg. Medical Domain)


Architecture/System components


Challenges


Future directions


Reference



3


What is Semantics vs.
Semantic Web?



Semantics:


focuses on the relation between signifiers, such as words, phrases, signs and
symbols, and what they stand for.


“Basically, what the sentence really means”



Semantic Web:


The Semantic Web is data that enables machines to understand the meaning
of information on the Internet. It extends the network human
-
readable web
pages by inserting machine
-
readable metadata and how they are related to
each other. This enables automated agents to access the Web more
intelligently and perform tasks on behalf of the user



Using metadata to add (and extract) meaning”



l




4

Understanding the Semantics


Computers don’t understand the website they are
showing to us


They may understand the syntax, the semantics is not.


If computers can understand what we are looking for,
they can help us search better


From passively helping us, to actively helping us


Understanding the meaning behind the Webpage



l




5


What will Semantic Web
Understand?


The computer will learn them and understand
how they interact with each other.


Person


Place


Event


Things




Currently it is dependant on Keywords


6


History of Semantic Web



1989 by Sir Tim Berners
-
Lee

7


Current Research Areas


Representation


Reasoning Engines


Query Languages





Web Ontology Language
-

RDFS

RDF Schema


RDF is a data model for objects and relations between
them


RDF Schema (RDFS) is a vocabulary description
language
based on XML and logic programming.


Describes properties and classes of RDF resources


Provides semantics for generalization hierarchies of
properties and classes

Web Ontology Language
-

OWL

OWL


A

richer ontology language
based on description logic.
More expressive languages than RDF Schema.


Is the current Web standard


Relations between classes


e.g., disjointness


Cardinality restrictions


e.g. “exactly one”


Boolean expressions


Richer typing of properties


characteristics of properties (e.g., symmetry)


Web Ontology Language
-

DAML+OIL


DAML was funded by US government. US Defense
Advanced Research Project Agency launched the
D
ARPA
A
gent
M
arkup
L
anguage to make web content
more accessible.


Ontology Inference Layer (OIL), a description logic.


DAML+OIL


developed by a joint committee from the US and the European
Union (IST) in the context of DAML, a DARPA project for
allowing semantic interoperability in XML.


DAML+OIL is built on RDF(S), extends with arbitrary data types
from XML Schema type system.


More Ontology Languages


SHOE


Simple HTML Ontology Extension
(University of Maryland)


OML


Ontology Markup Language (University
of Washington) is partially based on SHOE,
provides serialization of SHOE


XOL


Ontology Exchange Language (US
bioinformatics community) for exchange of
ontology definitions among different software
systems.

Expressiveness of languages


XOL


RDFS


SHOE


OML


OIL


DAML+OIL

Heavyweight

Querying Language


SPARQL


SPARQL is an RDF query language.

JENA Semantic Web Toolkit


Create and populate RDF data models in
Java applications.


Persist them to database


Query the models programmatically


Interface with Ontology engine


Interface with SPARQL

Reasoning Engine



Pellet OWL Reasoner for Java


An open
-
source Java based OWL DL reasoner.


Dual licensing model


Open source applications
-

can be used under the
terms of AGPL version 3 license


Closed source/commercial applications


can be used
under alternative license terms.


Execute SPARQL


http://clarkparsia.com/pellet/

16


Applications




Medical Domain


Product recommendations in eCommerce?





A Medical Information
Management System

(MIMS)


Uses Semantic Web Technologies


Diagnosis method of “dementia”


Data items are changed as research
progress.


Motivation for MIMS

Challenges faced by Medical researchers in developing a method to
diagnose dementia


Spends lot of time analyzing medical data in:


questionnaire survey,


metadata,


MMSE (Mini
-
Mental State Examination) data,


MRI (Magnetic Resonance Imaging) data,


MEG (Magnetoencephalography) data, and


physical checkup data. Some of them are saved as JPEG format,
DICOM (Digital Imaging and Communications in Medicine) format [1],


Microsoft Excel format, Microsoft


PowerPoint format and


Adobe PDF format.



Stored in network file folders; not easy to retrieve



Complicated retrieval needs


MMSE> 20, how to retrieve this info?



Semantic Web


The solution


A major hurdle of the Semantic Web is the
creation of metadata.


The metadata creation is a tedious and
laborious process for the researchers.


However, it is vital for articulating the
medical data.

MIMS


Essential components


(1) A pluggable metadata extractor


Metadata extraction mechanism that given as a plug
-
in which
automatically generates a set of metadata from medical data files.


(2) An RDFView


A semantic Web retrieval mechanism which provides Web
application programming interfaces created from SPARQL
templates dynamically.


(3) A representation mechanism

A method to display a result of the semantic Web retrieval service.

MIMS

(Contd…)

An overview of RDFView

An RDF model for DICOM image files

An overview of the pluggable
metadata extractor

The basic metadata schema
using RDF graph

Overview of MIMS

Many other medical applications


SemMed

Applying Semantic Web to Medical Recommendation Systems



Patient
-
Oriented Systems

24


Architecture/System components



XML layer


Syntactic basis


RDF layer


RDF basic data model for facts


RDF Schema


Ontology layer


More expressive languages


Current Web standard: OWL



Logic layer



enhance ontology languages
further


application
-
specific declarative
knowledge




Proof layer


Proof generation, exchange,
validation


Trust layer


Digital signatures


recommendations, rating agencies.


Semantic Web Architecture


A complete database architecture, not only an application program.


Semantic web architecture combines a two
-
step process.


First, a Semantic Web database is created from unstructured text
documents.


And, then Semantic Web applications run on the Semantic Web
database; not the original source documents.


The Semantic Web architecture is created by first converting text
files to XML


Then analysis performed around these with a semantic processor.
This process understands the meaning of the words and grammar of
the sentence, and also the semantic relationships of the context.


These meanings and relationships are then stored in a Semantic
web database. [13]

26


Semantic Web Challenges



Classification of semantic web challenges:



Implementation


Usage


General



Semantic Web Challenges

27

General Challenges






Multilingualism


to implement the same
page in several languages with or without
caching.


Social Hurdles


acceptability at
management level. Treated as another
mumbo jumbo.


28

Execution Challenges



Visualization


Scalability


Services


Trust



29


Implementation challenges




Ontological Modeling


Ontologies are key elements of semantic web.


Since these are extensions into a different domain of
knowledge, creation and development becomes
extremely hard.


Manual construction of Ontologies is a time
consuming operation.


30

Implementation Challenges



Ontology Engineering



Construction


Alignment


Merging


31

Implementation Challenges



Annotations and Metadata



Annotation of web contents is difficult.


Though it exists, it is not consistent.


32

Construction Challenges


Identifying and categorizing hierarchies of
concepts.


Categorizing documents.


Classifying concepts.


Finding non
-
taxonomic relationship
between documents.


Finding interrelated terms.

33

Mapping Challenges




Finding proper metrics for mapping.



Creating new classes.

34

Merging Challenges



Finding proper candidates for merging.



Finding proper metrics for merging once
the candidates are identified.

35

36


Future Directions




“Semantic Web”


Used in Facebook for establishing relationships and
connections.


Used in Learning Management Systems to classify
documents uniformly across all LMS.


Use in Medical record management systems.


More usages envisaged.




37

References


1.
John Davies, “Semantic Web” presentation
http://www.keapro.net/sekt/SemWebTutorialGeneralJD.ppt

2.
N. Henze and D. Krause, “Semantic Web Introduction”,
http://www.kbs.uni
-
hannover.de/Lehre/semweb06/all_in_one.pdf

3.
Charla Woodbury, David Embley, “Family History Research on the Semantic Web: Building a Semantic
Prototype for Danish Research”,
http://fht.byu.edu/prev_workshops/workshop05/FHTCD/session1/s1
-
CharlaWoodbury_SemanticWeb.pdf

4.
Deborah L. McGinnes et al, “DAML+OIL: An Ontology Language for the Semantic Web”, IEEE Intelligent
Systems Journal, 2002

5.
Asuncion Gomez
-
Perez and oscar Corcho, “Ontology Languages for the Semantic Web”, IEEE Intelligent
Systems Journal, 2002

6.
Mahmoud Barhamgi et al, “A framework for Data and Web Services semantic mediation in Peer
-
to
-
Peer based
Medical Information System”, 19
th

IEEE Symposium on Computer
-
Based Medical Systems (CBMS’06), 2006

7.
Masaharu Hayashi et al, “A Medical Information Management System Using the Semantic Web Technology”,
Fourth International Conference on Networked Computing and Advanced Information Management, 2008

8.
MohammadReza Keyvanpour et al, “Comparative Classification of Semantic Web Challenges and Data Mining
Techniques”, 2009 International Conference on Web Information Systems and Mining


9.
Kees van der Sluijs, “Semantic Web Applications” presentation,
http://wwwis.win.tue.nl/~ksluijs/material/wis
-
semwebapplications
-
class.ppt

10.
Grigoris Antonious, Frank van Harmelen, “A Semantic Web Primer”, 2008,
http://kianian.com/userfiles/semantic
-
web
-
primer.pdf

11.
Philip McCarthy, “Introduction to Jena”,
https://www.ibm.com/developerworks/java/library/j
-
jena/

12.
SPARQL Query Language for RDF,
http://www.w3.org/TR/rdf
-
sparql
-
query/

13.
Semantic Web Architecture and Applications, http://www.oss.net