Transcript Semantic Web - School of Computing and Engineering

CS690L
Semantic Web and Knowledge Discovery:
Concept, Technologies, Tools
Yugi Lee
STB #555
(816) 235-5932
[email protected]
www.sice.umkc.edu/~leeyu
1
CS690L - Lecture 1
General stuff
• Class hours: R 5:30 – 8:15pm FH310
• Office hours:
– Yugi Lee: T/TH 3:15-4:15 or by appointment
• Class page: www.sice.umkc.edu/~leeyu
– Lecture notes will be available in advance
– Mailing List
• Suggested Prerequisites
 CS551 (Advanced Software Engineering)
 Middleware Technology, component-based model (UML),
Object-Oriented programming language (Java, C++)
CS690L - Lecture 1
2
Reference books
•
•
•
•
•
•
•
•
•
XML Databases and the Semantic Web by Bhavani Thuraisingham, Bhavani Thuraisingha
CRC Press; ISBN: 0849310318 ; 1st edition (March 27, 2002)
Ontologies: A Silver Bullet for Knowledge Management and Electronic Commerce
by Dieter Fensel Springer Verlag; ISBN: 3540416021 ; 1st edition (August 15, 2001)
Knowledge Representation: Logical, Philosophical, and Computational Foundations
by John F. Sowa, David Dietz Brooks/Cole Pub Co; ISBN: 0534949657 ; 1 edition
(August 17, 1999)
Conceptual Spaces: The Geometry of Thought
by Peter Gardenfors MIT Press; ISBN: 0262071991 ; (March 20, 2000)
Internet Based Workflow Management: Towards a Semantic Web
by Dan C. Marinescu John Wiley & Sons; ISBN: 0471439622 ; 1st edition (April 5, 2002)
Spinning the Semantic Web: Bringing the World Wide Web to Its Full Potential
by Dieter Fensel (Editor), Wolfgang Wahlster (Editor), Henry Lieberman (Editor) MIT
Press; ISBN: 0262062321 ; (November 15, 2002)
Data Mining: Concepts and Techniques by Jiawei Han and Micheline Kamber, ISBN: 155860-489-8
Data Mining: Practical Machine Learning Tools and Techniques with Java Implementation
by Ian H. Witten and Eibe Frank; ISBN: 1-55860-552-5 Morgan Kaufamann Publishers
Further material will be made available through handouts in class and through pointers to
relevant Web pages.
CS690L - Lecture 1
3
Motivation
• The World-Wide Web has revolutionized almost every
information-related activity that people have been pursuing
throughout civilized history.
• 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.
• If the vision of a Semantic Web becomes a reality, this would
constitute a second revolution that would impact how we are
living our everyday lives.
• If we succeed in making a step towards the Semantic Web then
our work would have an impact on business, government,
education, research and many other domains.
CS690L - Lecture 1
4
Course Objectives
• Introduce the concept of Semantic Web, including its relationship to
Ontology and knowledge retrieval, the importance of Semantic Web
representation, including XML, RDF, DAML+OIL, and OWL and their
tools;
• Present an introduction to theoretical and practical aspects of
Knowledge Discovery: understanding various machine learning and
data mining algorithms and techniques for evaluating the performance
of the algorithms (classification, association, clustering, statistical
pattern recognition, neural networks, Bayesian learning, genetic
algorithms);
• Throughout this course, extensive hands-on exercises in problems in
Semantic Web and Knowledge Discovery with various knowledge
retrieval tools, will provide students with a better understanding of the
paradigm for Knowledge Discovery in Semantic Web.
CS690L - Lecture 1
5
Course Requirements
• A research-oriented graduate course,
– A substantial portion of the quarter will be devoted to
student presentations of techniques and research papers
– Students will be expected to select a problem area in
distributed computing architecture and prepare an
intensive presentation covering the methods and
framework commonly employed to address their
problem.
– A research paper on the application of a particular
middleware architecture is also an acceptable topic.
CS690L - Lecture 1
6
Course Requirements
• Discussion/Presentation:
– 3 – 4 Workshops (one or two presentations/ workshop)
– The lecture/discussions are designed to be highly
participatory.
– Participation will include such activities as group
discussions of topics through workshops; discussions with
faculty and student groups on topics, research, and/or
application problems;
– Short presentations on relevant papers and project results;
– Critiques of resource material, software, and other things
related to distributed component architecture.
CS690L - Lecture 1
7
Course Requirements
 Critical Reading/Thinking:
 30 ~ 50 research papers
 Students are required to read and assimilate information
from the readings beyond the material covered in class.
 Throughout the semester, papers and chapters of the
reference books will be read and discussed.
CS690L - Lecture 1
8
Course Requirements
 Analytical Writing:
 2 - 3 technical reports and one research paper
 Students are asked to think critically and reason about
information presented in the textbooks or papers. For
example, a paper assignment might ask how different
frameworks we are studying compare, or how existing
technology, like the Web will evolve in the context of
component software.
 This critical evaluation requires that students offer their
own understanding of the significance of what students
have learned.
CS690L - Lecture 1
9
Course Requirements
 Discovery (Self-guided) Learning:
 One or two areas (Semantic Web or Data Mining
specialized expertise)
 The course project will require independent research
and programming, and students are expected to be able
to demonstrate ability of this kind.
CS690L - Lecture 1
10
Assessment
• Group Project
40%
 Projects
 Group Activities
 Individual Work
60%
– Papers
– Presentation & Discussion
40%
20%
Both components must be passed in order to pass the course.
CS690L - Lecture 1
11
Group Projects
• Teams of maximum 4 members, development of a
component-based system.
• The overall assignment will be split into several
steps that will be marked individually.





Project 1 : Project proposal
Project 2: System design
Project 3: System implementation(1)
Project 4: System implementation(2)
Project 5: Documentation & final package
CS690L - Lecture 1
12
Group Projects
“Intelligent Knowledge Discovery Tool for Semantic Web”
– Building system followed by component-based design and
programming (Pluggable components)
– Hierarchical Management Structure (Chief Architect,
Intermediate Architect, Subgroups, etc)
– Incremental outcomes going through Component-based software
development process, such as requirement analysis, design,
implementation, testing, and integration.
– Object-Oriented Specification/Design (UML/Together), Design
patterns, styles, Object Framework building using XML, Web
Services or .NET.
CS690L - Lecture 1
13
Research Paper
• Goal: A research paper (Individual work)
• Tentative paper topics:
– Paper 1: Semantic Web
– Paper 2: Knowledge Discovery (Data Mining, Machine
Learning, etc)
– Paper 3: Knowledge Discovery for Semantic Web
CS690L - Lecture 1
14
Presentation & Discussion
Each student will be participated in three types of
presentations & discussion
 Workshop
 Paper
 Project
 Small group presentation/discussion
CS690L - Lecture 1
15
Contents of Lecture
Workshop 1: Semantic Web
–
–
–
–
Ontology and Ontology maintenance,
Interoperability, Integration, Composition
Web Services
Representation
•
•
•
•
XML and XML Schema
RDF and RDF Schema
DAML+OIL, DAML-S
Others (OWL, SHOE, RuleML, etc)
CS690L - Lecture 1
16
Contents of Lecture
Workshop 2: Knowledge Discovery
–
–
Data Mining and Machine Learning: Prediction,
Decision Supporting and Knowledge Discovery
Approaches
–
–
–
–
–
Classification
Association
Clustering
Web Mining
Statistical pattern recognition, neural networks,
Bayesian learning, genetic algorithms
CS690L - Lecture 1
17
Contents of Lecture
Workshop 3: Knowledge Discovery in Semantic Web
– Representation
– Algorithms
– Tools
– Killer Application
CS690L - Lecture 1
18