SPICE An Ontology Based Web Application

1
SPICE! An Ontology Based Web Application

• By
• Angela Maduko
• and
• Felicia Jones
• Final Presentation For CSCI8350 Enterprise
  Integration

2
Motivation

• In most cases when a user is performing a search,
  he is not only looking for a specific information
  but also wants to learn about related information
  or related categories
• Users searching for information in a particular
  domain would want to be assured that the
  information he obtains has a certain degree of
  validity within that domain

3
Introduction

• SPICE! is a web-based tool that integrates a
  number of
• different semantic toolkits and builds upon their
  basic
• functionality to provide search capabilities for
  the LSDIS
• Library website as well as retrieving related
  results from
• two other sites. SPICE! also creates RDF pages
  for the
• publications of the Library.

4
Overview

• Tools Integrated
• The Ontology
• System Architecture
• Design Issues
• Conclusion

5
Tools Used

• Protégé
• Ontojava
• Jakarta-Tomcat Apache Server

6
Protégé

• Protégé is a tool that facilitates
• The construction of knowledge bases in a
  principled fashion from reusable components
• The construction of domain ontologies
• and a lot more

7
Ontojava

• Ontojava is a cross compiler that automatically
  maps RDF facts, an ontology expressed in RDFS and
  rules and constraints represented in RuleML into
  a main memory database written in Java

8
Jakarta-Tomcat Apache Server

• Tomcat is the servlet container that is used in
  the official Reference Implementation for the
  Java Servlet and JavaServer Pages (JSP)
  technologies.
• Tomcat is developed in an open and participatory
  environment and released under the Apache
  SoftWare License.

9
SPICE! Ontology
10
(No Transcript)
11
(No Transcript)
12
System Architecture
13
System Architecture

• Protégé was used to create the ontology (RDFS and
  RDF format)
• The RDFS and RDF files output by Protégé serve as
  the input of OntoJava which are then converted
  into a Java object-oriented database
• An object oriented database is used because it
  provides for the ability of retaining the is-a
  relationships in the ontology
• eg. public class Book extends Publications

14
System Architecture

• The instances in the RDF files are stored in the
  database created by OntoJava
• The search Engine traverses the ontology each
  time a search is performed, in order to retrieve
  the best match for the users criterion
• Where the user does not specify any class type,
  the SPICE! performs an upwards traversal of the
  Ontology else, it performs a downwards traversal
  of the ontology

15
System Architecture

• In addition, the search criterion input by the
  user is forwarded to the Citeseer Research Index
  and DBLP Bibliography Server and their
  appropriate search results are also retrieved
• Updates to the LSDIS library WebPages is of
  course anticipated
• When a page is updated, the user enters the URL
  of the updated page in the update module

16
System Architecture

• Our Parser reads the html file, generates an RDF
  file for the page
• To input the new instances into the database,
  Ontojava is then automatically run again
• The Jakarta Tomcat server is also automatically
  shutdown and restarted in order that the updated
  java class files be detected

17
SPICE! INTERFACE
18
Design Issues

• SPICE! doesnt provide support for namespaces
• Multiple inheritance is not supported
• SPICE! is structured to work only with the
  current structure of the LSDIS Library WebPages

19
DEMO TIME

• SPICE!

20
References

• http//www.i-u.de/schools/eberhart/ontojava
• http//jakarta.apache.org
• http//protégé.stanford.edu
• http//lsdis.cs.uga.edu/lib/lib.html