Forum des doctorants (Nesrine)

MODULAR KNOWLEDGE BASE FOR
SEMANTIC WEB SEARCH
Academic Chair
in
Business Intelligence
Nesrine Ben MUSTAPHA, Marie-Aude AUFAURE
{Nesrine.ben-mustapha, marie-aude.aufaure}@ecp.fr
Context and Motivation
The Web bears the potential of being the world’s greatest
encyclopedic source the advances in automated information
extraction from Web pages give rise to “knowledge harvesting
from Web”.
Motivation: Harvesting knowledge according to users queries
and introducing Ontology modularization for knowledge
reuse and search over any domain area.
Issues:
Information extraction is time consuming from a large
scale corpora (the web)
Domain knowledge is dynamic and scalable
Modular Organization of a large knowledge base for
multiple domains
Objective: build a comprehensive machine-readable
knowledge base for semantic search
MODULAR KNOWLEDGE BASE FOR SEMANTIC WEB SEARCH
Example of modular domain ontology for E-commerce
High topic
Multi-Layer Ontology Warehouse
E-commerce
G
R
A
N
U
L
A
R
I
T
Y
Topic layer
Customer
module
Person
Module
Payment
module
Product
module
Domain Ontology = module network
Customer Module
Person
Module
Issued by
Person
Person
name
Customer
client
has
Indexing
Settled
_Via
Is_a
has
Order
purchase order
has
country
Ressources
Payment Module
Low
Payment
Method
job
Index
Index
ONTOLOGY MODULE-BASED SEMANTIC
SEARCH
Forum page
Internet
Ontology Modules
Book shopping page
page
Publications
Ben Mustapha N., Aufaure MA , Baazaoui Zghal H. and and Ben Ghezala H., Contextual Ontology Module Learning from Web Snippets and Past User Queries , 15-th Annual KES Conference Kaiserslautern, Germany, 12-14 September 2011
Ben Mustapha N., Baazaoui Zghal H., Aufaure MA and Ben Ghezala H. Enhancing Semantic Search using Case-Based Modular Ontology , the 25th Symposium On Applied Computing (SAC) , track on “The Semantic Web and Applications” (SWA), 22 –
26 March 2010, Sierre, Switzerland.