SYLLABUS MIS
6093-Independent Study in MIS-Data Mining
Instructor:
Dr. Richard S. Segall
Office:
216 Business Building
Phones:
972-3416 Extension 159 (Office) and
931-9642 (Home: Answering Machine available)
Class
Times: MIS 4093:
Other Classes:
&
Office Hours:
E-mail
Addresses: RSEGALL@MAIL.ASTATE.EDU (Office) and
RSSEGALL@AOL.COM (Home)
Course Web Site Address: http://business.astate.edu/econdsci/index.html & click on “Faculty” and “Richard Segall”
Text
Book Web Site Address: http://www-faculty.cs.uiuc.edu/~hanj/DM_Book.html
http://www-courses.cs.uiuc.edu/~cs497jh/papers/supplementarylist.htm (Cases for Han & Kamber)
Texts: 1.) Han & Kamber
(H&K): Data Mining: Concepts and Trends, Morgan Kaufman Publishers
(MKP)
ISBN 1-55860-489-8, Copyright 2001.
2.) Westphal
and Blaxton (W&B) Data Mining Solutions:
Methods and Tools for Solving Real World
Problems,
John Wiley & Sons, Inc., ISBN
0-471-25384-7, Copyright 1998.
3.) Additional handouts as distributed to class and/or made available on
course web site.
Prerequisites:
MIS 2403 or equivalent or knowledge of database
Proposed
Catalogue Description of Course: Concepts in data mining
techniques with emphasis on applying the concepts to obtain solutions for
different business situations.
Overview
of Course: One of the purposes of this course is to help you
become more aware of the concepts, trends and tool available for data mining in
the current Information Systems world. The course will discuss the fundamentals
of data mining concepts and techniques. This course has also primary objectives
of exposing students to some of the software available for data mining
solutions. Through class lectures, homework and student team presentations of
current relevant literature either distributed in class or available on the
World Wide Web, and a semester team project the students will become familiar
with this new area of information systems.
Outline and Course Schedule:
Chapter 1: Introduction
1.1
What motivated data mining? Why is it important?
1.2
So, what is data mining?
1.3
Data mining-on what kind of data?
1.4
Data mining functionalities-what kinds of patterns can be mined?
1.5
Are all of the patterns interesting?
1.6
Classification of data mining systems
1.7
Major issues in data mining
Chapter 2: Data Warehouse and OLAP Technology for Data Mining
2.1
What is a data warehouse?
2.2
A multidimensional data model
2.3
Data warehouse architecture
2.4
Data warehouse implementation
2.5
Further development of data cube technology
2.6
From data warehousing to data mining
Chapter 3: Data Preparation
3.1 Why
preprocess the data?
3.2
Data cleaning
3.3
Data integration and transformation
3.4
Data reduction
3.5 Discretization and concept hierarchy generation
Chapter 4: Data Mining Primitives, Languages, and System Architectures
4.1
Data mining primitives: what defines a data mining task?
4.2 A
data mining query language
4.3
Designing graphical user interfaces based on a data mining query language
4.4
Architecture of data mining systems
Chapter 5: Concept Description: Characterization and Comparison
5.1
What is concept description?
5.2
Data generalization and summarization-based characterization
5.2.1
Attribute-oriented induction
5.3
Analytical characterization: analysis of attribute relevance
5.4
Mining class comparisons: discriminating between different classes
5.5
Mining descriptive statistical measures in large databases
5.6
Discussion
Chapter 6: Mining Association Rules in Large Databases
6.1
Association rule mining
6.2
Mining single-dimensional Boolean association rules from transactional
databases
6.3
Mining multilevel association rules from transaction databases
6.4
Mining multidimensional association rules from relational databases and data
warehouses
6.5
From association mining to correlation analysis
6.6
Constraint-based association mining
Chapter 7: Classification and Prediction
7.1
What is classification? What is prediction?
7.2
Issues regarding classification and prediction
7.3
Classification by decision tree induction
7.4
Bayesian classification
7.5
Classification by back-propagation
7.6
Classification based on concepts from association rule mining
7.7
Other classification methods
7.8
Prediction
7.9
Classifier accuracy
Chapter 8: Cluster Analysis
8.1
What is cluster analysis?
8.2
Types of data in clustering analysis
8.3
A categorization of major clustering methods
8.4
Partitioning methods
8.5
Hierarchical methods
8.6 Density-based methods
8.7
Grid-based methods
8.8
Model-based clustering methods
8.9
Outlier analysis
Chapter 9: Mining Complex Types of Data
9.1
Multidimensional analysis and descriptive mining of complex data objects
9.2
Mining spatial databases
9.3
Mining multimedia databases
9.4
Mining time-series and sequence data
9.5
Mining text databases
9.6
Mining the World-Wide Web
9.6.4
Web usage mining
Chapter 10: Data Mining Applications and Trends in Data Mining
10.l
Data mining applications
10.2
Data mining system products and research prototypes
10.3
Additional themes on data mining
10.4
Social impacts of data mining
10.5
Trends in data mining
Appendix A An Introduction to Microsoft's OLE DB for Data Mining
Appendix B An Introduction to DBMiner
Readings to be assigned from text by Westphal
& Blaxton (W&B) to be selected from the
following:
Section I: Defining the Data
Mining Approach
Chapter 1: What is Data
Mining?
Chapter 2: Understanding
Data Mining
Chapter 3: Defining the
Problems to be Solved
Section II: Data Preparation and Analysis
Chapter 4: Accessing and
Preparing the Data
Chapter 5: Visual
Methods for Analyzing Data
Chapter 6: Non-Visual
Analytical Methods
Section III: Assessing Data Mining Tools and Technologies
Chapter 7: Link Analysis
Tools
Chapter 8: Landscape
Visualization Tools
Chapter 9: Quantitative
Data Mining Tools
Chapter 10: Future
Trends in Data Mining
Section IV: Case Studies (Student Presentation are also to be selected from these Chapters as well
as Readings List cited earlier.)
Chapter 11: Mapping the
Human Genome
Chapter 12:
Telecommunication Services
Chapter 13: Banking and
Finance
Chapter 14: Retail Data
Mining
Chapter 15: Financial
Market Data Mining
Chapter 16: Money
Laundering and Other Financial Crimes
Since
this is a Graduate course per AACSB Accreditation standards Graduate students
are to be
evaluated
on different higher requirement than undergraduate students enrolled for same
Course
Hence
Graduate Students will in addition to the above will be required to write a ten
(10) page
paper*
(excluding additional pages as needed for references, tables and figures) on an
application(s) of Data Mining of their selection. The grade for this paper will
be a letter grade that will be included into the Written Homework assignments
scores out of a possible 200 points.
*
Please refer to graduate school catalogue for policies regarding plagiarism.
All citations are expected to be cited in this paper.
Grading
Policy:
The course grade will be determined by the following:
Three
(3) Exams (In-class & Take-Home parts) and any quizzes = 45%
Written
Homework Assignments & ten (10) page Paper = 22%
Team
Project and Class Presentations =
18%
Final
Exam (In-class & Take-Home part) =
15%
Course
Policies:
1.) Portions
of the exams and/or final exam may be composed of take-home portions because
the nature of the course requires use of computers, and thus cannot be
completed in-class. It is expected that you DO YOUR OWN WORK on these, as well as other assigned homeworks. Copying
homework or submitting identical computer files or printouts from other
student(s) is not allowed!!!
Similarly
submitting homework or take-home exam answers that are entirely copied from the
text verbatim is
considered as plagiarism. Submitting verbatim text of any webpages without citations or from other software
modules on take-home exams or homeworks or semester team project is also
considered as plagiarism. Victims of collusion on assigned homeworks and take-home exam portions and plagiarism as
described above will be penalized!!! You are responsible for knowing contents
of “ASU Student Handbook” and its stated “Academic Integrity Policies.”
2.) Although
no explicit policy or statement regarding class attendance is written in the ASU
2002-2003 Graduate Bulletin, it is expected that graduate students attend
all classes and abide to policies similar to that stated on page 41 of the
ASU 2001-2002 Undergraduate Bulletin:
“Students should attend each lecture, recitation, and laboratory session
of every course on which they are enrolled. Students who miss a class
session should expect to make up missed work or receive a failing grade on
missed work. Make-up policy is at the discretion of the instructor.”
“Students enrolled in junior and senior level courses (numbered 3000 or
4000) will not be assigned a grade of F solely for failing to attend classes. However,
instructors shall set forth at the beginning of the semester their expectations
with regard to make-up policy for work missed, class participation, and other
factors that may influence course grades.”
3.) Exceptions
to this rule as stated above will be made for excusable absences as documented in writing such as medical
excuses, death in family, etc. In summary, you are responsible for everything
discussed or presented in class regardless of you were in attendance or not. Attendance
is essential especially when Student Presentations are to be made, as questions
need to be directed to presenters.
Those with perfect
attendance will be given the benefit of the doubt on borderline grade
cases at the
end of the
term.(example: between 89.99 and 90.00 as shown in item (13.) below.)
4.) If
you come to class late and find the classroom empty, do not assume that class
is canceled. Assume that the class is meeting in one of the other computer
rooms for hands-on instruction or some other classroom. You are expected to
check the computer room(s) to locate the class and announcements on blackboard
or classroom doors indicating location of the class.
Hence similar to
university policy stated on page 41 of the of the ASU 2001-2002 Undergraduate
Bulletin, class attendance will be
taken daily. Those
with perfect attendance will be given the benefit of the doubt on
borderline grade cases at the end of the term. (example: between 89.99 and
90.00 as shown in item (13.) below.)
5.) Class Attendance will be taken regularly
with sign-up sheets to be passed around the room. It
is your responsibility to see that the daily attendance sheets are signed! Class attendance sheets may be passed
around the class twice during the class period, i.e. once at start of
class and once later in class. In both of these cases, signing the attendance sheet only once
would NOT constitute full attendance for that class!!!
6.) Point deductions for the
overall class average earned for the term will be made according to the
following scale: of one-half point deduction for each unexcused class
absence beyond first:
Perfect Attendance :
0
Points
One Absence
1.0 Points
Two Absences:
2.0 Points
Three Absences:
3.0 Points
Four Absences:
4.0 Points
Five Absences: 5.0 Points
Six Absences:
6.0 Points
Etc. ……………………………………………….
Example:
90.53 overall course average with 2 absences yields adjusted overall course
average of 89.03 and B course grade.
7.) All
assignments submitted for this course are expected to be word-processed. All
assignments for
this course having diagrams are expected
to be computer generated (i.e. NOT hand drawn!).
Assignments not following this
requirement will either be not accepted or receive severe point
deductions. As a general rule, assignments not
word-processed will not be accepted for grading!!!
Software available in computer labs include: MS Word with its toolbar symbols, and for
presentations: MS PowerPoint, WinEdge, Visio Professional (pre-made components useful in
Diagrams),
SmartDraw (for
self-made diagrams) downloadable from www.smartdraw,com ,
etc.
8.) All homework, take-home portions of
exams, and take-home-final portion are expected to be handed
in on time!!!
Late homework will not be accepted unless there is a valid excuse e.g.
illness. If is known
in advance
that you will be unable to attend class when homework or take-home exam is due,
you are
advised to submit in advance, or send by e-mail making copies for
self of any computer files, or FAX
hard
copies to the Department of Economics & Decision Sciences at (870)910-8187 with Dr. Segall
name
as addressee in order that it will arrive by due date.
The
instructor however will not be responsible for lost submission not
handed in with the class, e.g. e-mail sent
by
error without attachment files or sent to incorrect e-mail address, or fax
transmission doesn’t get
delivered to instructor, or placed in office door basket or under door
that was due in class time, etc. You are
advised to make back-up files of all homework and take-home exams
submitted.
9.) Every semester,
there are a few students who attempt to hand in the entire semester’s worth of missing
homeworks collectively at the end of the semester to be
graded for credit. THIS IS TOTALLY
UNACCEPTABLE!!!
Also due in part to the immense amount of homework papers that need to be
graded weekly for all of Dr. Segall’s classes
this is unfeasible to even be done.
10) Make-up exams or
quizzes will only be made for valid reason (e.g. illness, death in family) and
need to be made-up
as
soon as possible. For example, missed exam #1 with documented reason does not
constitute valid reason for
make-up exam #1 at end of semester when answers have already been
discussed in class. No make-up would be
allowed for such a described situation.
11.) Formula for Grade Determination for Homework Contribution to Grade:
= ((Total homework points earned)/(Total
homework points assigned)) X 22
12.) Formula
for "Approximate"* Grade
Determination for each Exam**
contribution to Grade:
= ((Total exam points earned)/(Total exam
points possible)) X 15**
*approximate
because any quizzes points are also added to "Total exam points
possible."
**Both in-class and
take-home parts
13.) A
straight curve is intended to be used for course grade determination:
90.00-100 A,
80.00-89.99 B,
70.00-79.99 C,
60.00-69.99 D,
0-59.99 F.
Since the quizzes and exams are primarily* intended
to be open-book and open notes, and take-home exams given with ample time, the
above described straight curve is intended to be strictly adhered to!!!
There may however be some closed book parts on matching of terms and their
corresponding definitions.
14.) “Double-rounding” will NOT be employed in computing averages of components of final
course
grade. That is, rounding of
overall Homework average to be added to rounding of overall average of
Exams will NOT be used
in computing final grade for the course!!! As indicated above, it is unlikely
that a grading curve will be employed at the end of the term, but would
be used only if deemed
appropriate. In this unlikely event, no curve will be employed for any
individual exam because of the
composite set of exam scores for the entire term must be considered
along with the other components.
If
a grading curve is used during the term, it will be constructed in such a way
that “clusters” of
scores will be grouped together to receive
the same letter grade.
PLEASE NOTE THAT NO GRADING CURVE WAS
EVER EMPLOYED WHEN ANY
COURSE WAS TAUGHT BY DR. SEGALL IN
PREVIOUS SEMESTERS AT ASU!!!
15.) It is impossible to discuss in class the
contents of all of the assigned readings for the course. However
you will be responsible for the contents
of all of the assigned readings for examinations and assignments.
16.)
You are responsible for everything posted on the course
web page and periodically checking the
course web site for
updated and recent postings at the hotlink of MIS 4093/MIS 6093 at web
site
address of:. http://business.astate.edu/econdsci/index.html
(Click on “Faculty” & “Segall”). You
are
responsible for printing out and bringing to class copies of the
PowerPoint slides used in lecture for your taking
of
class notes. When clicking on the MIS 4093/MIS 6093 hotlink additional hotlinks will be available
to click
on
such as by individual chapter for viewing and printing these lecture materials.
17.) The Final Exam most likely cannot be
comprehensive in nature due to immense amount of readings
and
topics covered in this course. Most likely the Final Exam would emphasize the
material from the
later part of the course. If points remain in writing Final Exam so that
no additional questions can be
written to complete total points required, a few review questions may be
included. The exact contents
of
the Final Exam will be announced in class prior to the Final Exam.
18.) Per University Regulations, no grades will be
given over the phone or by e-mail.
19.) Use of cell phones during lecture time and
exams will not be tolerated. Please be
considerate of your classmates and the instructor.
20.) Those students with disabilities are
responsible informing Dr. Segall of written
documentation of their special needs for exams.
21.) The instructor reserves the right to make
any necessary changes to the course syllabus as
stated, and would announce any necessary
changes in class at the appropriate time(s).
------------------------------------------------------------------------------------------------------------------------------------------------------------
RULES FOR CLASS TEAM
PRESENTATIONS:
1.
Student
Class Team Presentations will be based upon current events in Data Mining
available on web sites provided by Dr. Segall, or
Case Studies for the Han and Kambler text provided by
the publisher at text web site of
http://www-courses.cs.uiuc.edu/~cs497jh/papers/supplementarylist.htm, or Case Studies contained
within Chapters
11-16 of Westphal & Blaxton,
or handouts provided by Dr. Segall. One of the
purposes of these
Student Team presentations is to provide more in-depth insight into the
applications of Data
Mining as Case Studies, insight into the state-of-the art of Data Mining as
provided by current
web postings, and to relate these to the respective chapter(s) in our Texts we
are discussing in
lecture.
2.
Students
are expected to do two (2) team presentations for the semester. If time permits
which depends on class size, extra credit will be awarded for any extra
presentations as points to be added to the total homework points earned.
3.
Teams
of presenters are to be formed by students selecting their teammates as a first
priority. Sign-up notices will be written on Class Attendance Sheets. Dr. Segall will then assign those students as needed to form
complete teams.
4.
Sign-up
for each student’s second presentation can not be made until a complete
cycle has been made of the class of the complete set of first presentations.
5.
Each
team member is to speak for 5 minutes each. Hence total time for each team will
be equal to:
(number of
presenters) times (5 minutes per presenter) + 5 minutes for Questions and
Answers from
audience.
6.
The
contents of each presentation should consist of:
(a.)
a
summary of the highlights of the Case Study or material to be presented.
(b.)
Indication
of how the Case Study or article relates to the Chapter(s) in Han & Kamber being discussed in lecture. Some of the case studies
will be from chapters 11 to 16 of text Westphal &
Blaxton.
(c.)
Potential
Take-Home Exam questions as discussed below. The set of questions for the
entire team should preferably be presented at the end of the presentation.
However for some of the in-depth case studies it may be more appropriate to
present each presenter’s question at the end of his or her’s
portion of the team’s presentation. The potential exam questions should either
be numbered consecutively or labeled as author’s questions, e.g. “Laurie’s
question”.
7.
Each
presenter must present one (1) essay or problem (e.g. table or figure) question
per presenter based on the material of their class presentation each of
which would be a potential question on the Take-Home part of an Exam. That is,
the potential Take-Home exam questions should be focused on the contents of the
presentation. The potential exam question can also relate to material in Han
& Kamber or Westphal
& Blaxton, but still must pertain to the contents
of their presentation. The potential exam questions can be essay, create a
table, etc. but NOT short answer (e.g. NOT T/F or MC or fill-in blank).
That is, short answers of five (5) or six (6) words would be totally
unacceptable as potential Take-Home Exam questions.
8.
Only
if any class presentations are based on web sites no longer available on web or
current literature not available on the web, will Dr. Segall
then distributed copies of the appropriate materials to each of the team
members prior to each team's class presentation.
9.
Presentations
are expected be made in Microsoft PowerPoint. Each team is required to give Dr.
Segall a diskette of the team presentation during the
class of presentation for posting on the course web page. This diskette is NOT
to have the answers to the proposed student Exam questions, even though the
student presentation given in class may include slides with acceptable answers.
10.
Each
presenter will be given a “Team Presentation Peer Evaluation Form” to complete
on the day of his or her Team Presentation. In this form, each presenter will
evaluate and grade the performance and contribution of the OTHER team members
of their team, i.e. they will be grading everyone else but themselves. These
forms are to be completed and handed in to Dr. Segall
by the end of the class on which the team presentation was given, and will be
held in strict confidence. Only letter grades of A, B, C, D, and F will be used
on this form.
11.
Presentation
Grades will not be made available until the end of the semester
on distributed
summary sheets because your presentation
grade will not only be made relative to the
quality of your contribution to your
team’s presentation but also relative to the quality of
the presentations made by the other
teams.
12.
Only
letter grades will be assigned to presentations, where:
A+=100, A=95, A-=90, B=85, B-=80, C=75,
C-=70, D=65, D-=60, F+=50, F=25, F-=0.
That is,
presentation grades such as 83, 91, etc. would not be possible.
SOURCES FOR SEMESTER TEAM PROJECT
CD-ROM of Westphal and Blaxton text has the
following software modules available for download. See Appendix “What’s on CD-ROM” on
pages 597-602 for descriptions of each of these software modules:
-
Analyst’s Notebook
-
Cross Graphs
-
Daisy
-
DataVista