Transcript Summary of Knowledge Discovery for Semantic Web

Summary of
Knowledge Discovery for Semantic
Web
Article by Dunja Mladenic, Marko Grobelnik,
Blaz Fortuna, and Miha Grcar, Chapter 3 in
Semantic Knowledge Management:
Integrating Ontology Management,
Knowledge Discovery, and Human Language
Technologies, Springer Verlag, Berlin, 2009,
21-35
Summary by Andrew Zitzelberger
What is the Semantic Web?
 The Semantic Web can be seen as mainly dealing
with the integration of many, already existing
ideas and technologies with the specific focus of
upgrading the existing nature of web-based
information systems to a more “semantic” oriented
nature.
What is Knowledge Discovery?
 Knowledge discovery can be defined as a process
which aims at the extraction of interesting (nontrivial, implicit, previously unknown, and
potentially useful) information from data in large
databases.
How Does Knowledge Discovery Help Us?
 Ontology Construction
 Domain understanding (what is the area we are dealing with?)


Data understanding (what is the available data and how is it
related?)


Information Retrieval
Machine Learning and Data Mining
Task definition (what to do with the data ?)
Ontology population
 Extending the ontology



Ontology learning (semi-automated process)
Ontology evaluation (estimate quality of solutions)


Gold Standards
Human refinement (iterate)
How Does Knowledge Discovery Help Us?
 Domain Knowledge
 Capture domain specifics

Track user’s search interests
 Dynamic Data
 How does data change over time?

Data drift and visualization of data changes
 Multimodal and Multilingual Data
 Non-textual data

Pre-processing other forms of data into more useful
representations
Tools
 OntoClassify
 Used for ontology population
 OntoGen
 Used to edit topic ontologies
 SEKTbar
 Used to maintain dynamic user profiles

Creates an ontology to model the interests of the user in order to
highlight items of expected interest on the pages the user is
visiting.
SEKTbar
SEKTbar