CS-XXX Data Mining (Fall 2009)

chardfriendlyΤεχνίτη Νοημοσύνη και Ρομποτική

16 Οκτ 2013 (πριν από 3 χρόνια και 9 μήνες)

74 εμφανίσεις

CS
-
XXX
Data Mining

(Fall 200
9
)

Instructor: Yu
-
Chieh Wu (
吳毓傑
)


General information

This course introduces some of the central themes and techniques that have emerged
in
data mining related research topics, especially for supervised classification and
unsu
pervised clustering etc.

Techniqu
es covered will include
neural networks, support
vector machines, boostrapping, decision trees, association pattern mining, sequential
pattern mining, text mining, etc. Industrial topic is also welcome. Programming skill
is

a plus in this class. No need to implement special system or algorithm. The extra
point is magic for most people who are interested in propose novel ideas or real world
problems. Solutions should be rephrased and resolved for the special problems.


The c
ourse is lecture
-
based. Grading is based on a midterm examination,
team
project
,
extra point, and several homework assignments
.
In front stage,
all lectures
will be instructed before mid
-
term.
At final stage, all classmates should propose their
topics or p
roposal.
The course is suitable for computer science or engineering students
who are preparing doing related research issues.


Teaching Method

Traditional lecturing with PPT slides (Major) and electric blackboard

During class, several out
-
of
-
topic works w
ill be discussed.

Industrial topic or problem is very welcome.


Administrative



Time
:
07
:
15

p
.m.


9
:
45

p.m.
Fri

(18 weeks)



Place
: TBA



Instructor
: Yu
-
Chieh Wu

o

Office: TBA

o

Tel:
TBA

o

Email:
bcbb@db.csie.ncu.edu.tw

o

Office hours
:
TBA




TA
: TBA

o

Office hours
: TBA


Textbook



Introduction to Data Mining
, by
Pang
-
Ning Tan, Michael Steinbach, and Vipin
Kumar
,
Pearson International Edition, 2005.



http://www
-
users.cs.umn.edu/~kumar/dmbook/index.php

Refe
rences



Web Data Mining Exploring Hyperlinks, Contents and Usage Data
, B. Liu,
Springer, December, 2006
.



Data Mining:


Concepts and Techniques, J. Han and M. Kamber, Morgan
Kaufmann , 2000.


Prerequisites

Finding new and interesting problem is very importan
t, in particular for the real world
or industrial purpose.
Knowledge of basic probability and statistics is a plus.

In
addition, this course will discuss
several

advanced prediction models.


Course Policy and Grading

Grading will be based on the following
weighting scheme,

o

Attendance: 10%

o

Mid
-
Exam:
2
0%

o

Team Project and Presentation:
5
0%

o

Homework:
3
0%

o

Extra:
2
0%


There are two homework assignments. Each of which is very important for this class.
DO NOT BE LATE!

The extra points are only available for student
s whose final score is below 70. Detail
instructions of how to earn the extra points will be announced before final project and
presentation.


Syllabus

Topic

Week

Hours

Assignment

Introduction

1

3



Week

1

Chapter 1: Introduction

1

3



Week

2

Chapter 2: Data

1
.5

4



Week

4

Chapter 3: Exploring Data

1.5

5



Week

5

Chapter 6: Association Analysis: Basic Concepts and Algorithms

2

6

Homework

Week

7

Chapter 7: Association Analysis: Advanced Concepts

2

6



Week

9

Chapter 4: Classification Basic Concepts, Decision Trees, and Model
Evaluation

2

6

Homework

Week

11

Chapter 5: Classification: Alternative Techniques

3

9



Team Project and Presentation Proposal

Mid
-
Term

Week

13

Chapter 8: Cluster Analys
is: Basic Concepts and Algorithms

1

3



Week

14

Chapter 9: Cluster Analysis: Additional Issues and Algorithms

1

3



Week

15

Presentation

1

3



Week

16

Presentation

1

3



Week

17

Presentation

1

3



Week

18

Presentation

1

3



Final Term Exam




Projec
t Topics:
(Not limited)
:


Evaluating Performance of Classifiers

Support Vector Machine (SVM)

Pattern Recognition Methods

Learning Strategy

Semi
-
supervised learning (classification with labeled and unlabeled data)

Classification for rare
-
class problems

Sequ
ence Prediction

Association Rules for Classification

Spatial Association Rule Mining

Temporal Association Rule Mining

Tree Mining

Sequential Association Rule Mining

Outlier Detection

Clustering



p.s.
Feel free to use the above topics as keywords to sea
rch related papers or reports
from Google.











:

歐亞出版社




:

資料探勘

1/E (TAN:INTRODUCTION TO DATA MINING 1/E)



號:

0080002



者:

施雅月




價:

650



購:

553

聯絡人:

邱傳龔

0922
-
900
-
152

服務電話:

02
-
89121188


02
-
89121188



真:

02
-
89121166


02
-
89121166



站:
http://www.eurasia.com.tw