CSCMPE 636 Advanced Data Mining

1
CS/CMPE 636 Advanced Data Mining

• Outline

2
Description

• Cover recent developments in some key areas of
  data mining
• Mining data streams
• Cluster analysis
• Web mining
• Prepare students for research work in data
  mining.
• Follow a lecture-discussion format where topics
  are introduced and techniques critically
  discussed. The majority of the material discussed
  will be derived from research publications.
  Students will be expected to read before coming
  to class and participate in the discussions.
• Emphasis will be placed on the design and
  implementation of efficient and scalable
  algorithms for data mining.
• The course project will require students to
  research, design, implement, and present their
  solution to a data mining problem.

3
Goals

• To expose key research areas in data mining
• To develop article comprehension and critical
  review skills
• To improve research and presentation quality for
  possible publication

4
After Taking this Course

• You should be able to
• comprehend and critically analyze data mining
  research
• design and implement data mining solutions
• write and publish articles

5
Prerequisites

• CS 536 Data Mining This course provides
  necessary concepts and foundations for CS 636
• Permission of instructor
• For those who have taken CS 535 (Machine
  Learning) and are motivated and willing to learn
  data mining basics on their own
• For any other super motivated person
• Passion for learning, research, and development

6
Grading

• Points distribution
• Project 35
• Quizzes 20
• Assignments 5
• Attendance and CP 5
• Exam 35

7
Policies (1)

• Quizzes
• Most quizzes will be announced a day or two in
  advance
• Unannounced quizzes are also possible
• Sharing
• No copying is allowed for assignments.
  Discussions are encouraged however, you must do
  and submit your own work
• Violators can face mark reduction and/or reported
  to Disciplinary Committee
• Plagiarism
• Do NOT pass someone elses work as yours! Write
  in your words and cite the reference. This
  applies to code as well.

8
Policies (2)

• Submission policy
• Submissions are due at the day and time specified
• Late penalties 1 day 10 2 day late 20
  not accepted after 2 days
• An extension will be granted only if there is a
  need and when requested several days in advance.
• Classroom behavior
• Maintain classroom sanctity by remaining quiet
  and attentive
• If you have a need to talk and gossip, please
  leave the classroom so as not to disturb others
• Dozing is allowed provided you do not snore load
  ?

9
Project

• Research, design, implement and evaluate a data
  mining algorithm
• You may choose a problem of your liking within
  the focus areas of this course (after
  consultation with me) or select one suggested by
  me
• Each of you must do the project independently
• Overview
• Literature search and annotated bibliography
• Research review
• Solution/algorithm design
• Implementation and evaluation
• Report and presentation
• Start thinking about the project now

10
Summarized Course Contents

• Review
• Mining data streams
• Data stream models
• Algorithms
• Intrusion detection
• Cluster analysis
• Similarity measures
• Algorithms for data streams and mixed-type
  datasets
• Web mining
• Intelligent information retrieval
• Newgroup mining
• Coverge and contents may vary according to the
  dynamics of the course

11
Course Material

• Required
• No required textbook
• Set of articles to be put in the course folder on
  COMMON drive
• Supplementary material
• Data Mining Introductory and Advanced Topics,
  Dunham, Pearson Education, 2003.
• Data Mining Concepts and Techniques, Han and
  Kamber, Morgan Kaufmann, 2001.
• Other resources
• Books in library
• Web

12
Course Web Site

• For announcements, lecture slides, handouts,
  assignments, quiz solutions, web resources
• http//suraj.lums.edu.pk/cs636w04/
• The resource page has links to information
  available on the Web. It is basically a meta-list
  for finding further information.

13
Other Stuff

• How to contact me?
• Office hours 10.00 to 12.00 MW (office 429)
• E-mail akarim_at_lums.edu.pk
• By appointment e-mail me for an appointment
  before coming
• Philosophy
• Knowledge cannot be taught it is learned.
• Be excited. That is the best way to learn. I
  cannot teach everything in class. Develop an
  inquisitive mind, ask questions, and go beyond
  what is required.
• I dont believe in strict grading. But there has
  to be a way of rewarding performance.

14
General Reference Books in LUMS Library (1)

• Data Mining Concepts, Models, Methods, and
  Algorithms, Mehmed Kantardzic, 006.3 K167D, 2003.
• Principles of Data Mining, Hand and Mannila,
  006.3 H236P, 2001.
• The elements of statistical learning data
  mining, inference, and prediction, Tervor Hastie,
  Robert Tibshirani and Jerome Friedman, 006.31
  H356E 2001.
• Data mining and uncertain reasoningan integrated
  approach, Zhengxin Chen, 006.321 C518D 2001.
• Graphical models methods for data analysis and
  mining, Christian Borgelt and Rudolf Kruse, 006.3
  B732G 2001.
• Information visualization in data mining and
  knowledge discovery, Usama Fayyad (ed.), 006.3
  I434 2002.
• Intelligent data warehousingfrom data
  preparation to data mining, Zhengxin Chen, 005.74
  C518I 2002.
• Machine learning and data miningmethods and
  applications, Michalski, Ryszard S., ed.Bratko,
  Ivan, ed.Kubat, Miroslav, ed., 006.31 M149 1999.
• Data Mining Practical Machine Learning Tools and
  Techniques with Java Implementations, Witten et
  al., Morgan Kaufmann, 006.3 W829D, 2000.

15
General Reference Books in LUMS Library (2)

• Machine Learning, Tom Mitchells, McGraw-Hill,
  1997.
• Managing and mining multimedia databases, Bhavani
  Thuraisingbam, 006.7 T536M 2001.
• Mastering data miningthe art and science of
  customer relationship management, J.A. Michael
  Berry and Gordon Linoff, 006.3 B534M 2000.
• Data mining explaineda manager's guide to
  customer-centric business intelligence, Rhonda
  Delmater and Monte Hancock, 006.3 D359D 2001.
• Data mining solutionsmethods and tools for
  solving real-world problems, Christopher Westphal
  and Teresa Blaxton, 006.3 W537D 1998.