You are on page 1of 6


Version 08/18/2016


Fall 2015, CAVC 351, 4:30-5:45
Instructor: George C. Runger,, BYENG 320, 480-884-0220, 480-965-3193
Office hours: TTh, 3:15-4:15, other times by appointment
TA: Kangwon Seo,
Required Textbook: Introduction to Data Mining, P.-N. Tan, M. Steinbach, V. Kumar, Pearson
Addison Wesley, 2006
Recommended Textbook: Data Mining: Practical Machine Learning Tools and Techniques (Third
Edition) I. H. Witten, E. Frank, M. A. Hall, Morgan Kaufmann, 2011, ISBN 978-0-12-374856-0
Reference Textbooks: Elements of Statistical Learning, T. Hastie, R. Tibshirani, J. Friedman, Springer,
NY, 2001; Machine Learning, by T. Mitchell, McGraw Hill, NY, 1997
About the books:
The book by Tan et al. presents from a computer science perspective. It provides the mechanics of
modelsthe operational details. It is our primary book to understand a model. The book by Hastie et al.
presents from a statistical perspective, assumes that the details are easily grasped and focuses on
relationships and insights into the models. It was used previously as the book for the course. The book by
Witten et al. provides fewer details than the other books, but it has additional comments and topics, and it
focuses on the WEKA software. Tan et al. is the required book, but it will be supplemented with
additional material.
About the course:
A survey course for topics from data mining and machine learning are presented. Advantages and
disadvantages of methods are discussed.
Emphasis is on data models and concepts, rather than inference.
Prerequisite background: A working knowledge of basic statistical methods. A formal course in
engineering statistics at the level of ASE 485 is the official prerequisite. A previous course in
empirical modeling such as regression analysis or designed experiments is recommended. Some
experience with matrix algebra.
Course operated through the MYASU Web site. Please submit all material through the Web sitenot
WEKA software will be used for examples. WEKA software is limited, but valuable to illustrate
concepts. WEKA is a free download. R software will also be demonstrated.
Course objectives:
Learn selected data mining models, objectives, steps, inputs, outputs, assumptions, validity,
advantages, disadvantages, and relationships of methods. The focus is on classification, but regression
is considered for some models.
Learn concepts such as problem types, over and under fitting, how to evaluate a model.
The field of data mining spans a large collection of different models. At the surface these look fragmented
and disconnectedwithout a route to solve a particular problem.
Pure operation of a software package does not provide insight into a model for a particular problem.
Instead, a roadmap (or guidance) is needed to develop solution strategies.
We build a basic understanding of the methods through calculations with a set of simple data sets.
These examples are provided as the homework exercises, and they illustrate the objectives, steps,
inputs, outputs of each model. This allows us to understand the assumptions, advantages,
disadvantages, and relationships of methods, and this provides insight into the role of the various

tools in a solution (guidance). This approach allows us to focus on the key characteristics of models.
We do not focus on software implementations of algorithms. We analyze large, but not massive, data
sets in the course to avoid computing details which can overwhelm concepts.

Your course grade is determined from
two mid-term exams (40%),
mini-projects completed in teams, including in class presentations, quizzes, etc. (for the in class
students only) (15%),
comprehensive final exam (30%),
individual final analysis project (15%).
Recommended homework exercises are provided. These are not submitted nor graded. There are a
minimum number of exercises and more should be worked as necessary to master the material.
All exams are open book and notes and emphasize the calculation and interpretation of modelbuilding concepts. Computers are not permitted. A calculator is needed.
Mini-projects completed in teams with a small report--approximately every two weeks.
An individual final project is used to apply techniques from the course on a larger data set. Here, the
learning is put to use as a sequence of steps is developed and implemented for a data set and problem
objective. In the project, when one attempts to develop and implement a model, the complexities of
an analysis can arise. Still, the modeling fundamentals provide the guidance for an effective solution.
Computer software is expected to be used in the project. The data set and problem objective for the
final project will be provided by the instructor.
You are required to know and follow ASU Academic Integrity Policiessee further information
A weighted average score from all requirements determines the final course grade. The following scale
provides a minimum for your course grade:
90-100, A; 80-89, B; 70-79, C; 60-69, D; below 60, E

Course Policies & Procedures

Technology Enhanced Course
This is a both a face-face and online course that requires attendance in face-face meetings and utilization
of online resources.
Communicating With the Instructor
This course uses a Blackboard discussion board called Hallway Conversations for general questions
about the course. Prior to posting a question, please check the syllabus, announcements, and existing
posts. If you do not find an answer, post your question. You are encouraged to respond to the questions
of your classmates. Email questions of a personal nature to your instructor or assigned TA or schedule an
appointment during office hours. You can expect a response within 48 hours.
Email and Internet
ASU email is an official means of communication among students, faculty, and staff
( Students are expected to read and act upon
email in a timely fashion. Students bear the responsibility of missed messages and should check their
ASU-assigned e-mail regularly.
All instructor correspondence will be sent to your ASU email account.
Campus Network or Blackboard Outage
When access to Blackboard is not available for an extended period of time (greater than one entire
evening) you can reasonably expect that the due date for assignments will be changed to the next day
(assignment still due by 11:59PM). If an outage occurs, it is expected that you will confirm that the
outage is with the University and not with your local internet service provider. To monitor the status of
campus networks and services, please visit the System Health Portal ( If a
system-wide ASU outage is NOT listed, you are responsible for contacting the ASU Help Desk to report
and troubleshoot the issue. By contacting the help desk, a request case number will be created for you,
which serves as an important documentation of your attempt to resolve any technical problems in a timely
fashion. You may be required to forward this documentation to your instructor.
Course Time Commitment
This three-credit 15 week course requires approximately 135 hours of work. Please expect to spend
around 9 hours each week preparing for and actively participating in this course.
Late or Missed Assignments
Notify the instructor BEFORE an assignment is due if an urgent situation arises and the assignment will
not be submitted on time. Published assignment due dates (Arizona Mountain Standard time) are firm.
Please follow the appropriate University policies to request accommodation for religious practices
( or to accommodate a missed assignment due to
University sanctioned activities (
Submitting Assignments
All assignments unless otherwise announced, MUST be submitted to the designated area of
Blackboard. Do not submit an assignment via other methods unless specifically directed.
Drop and Add Dates/Withdrawals
This course adheres to a set schedule and may be part of a sequenced program, therefore, there is a
limited timeline to drop or add the course ( Consult with your
advisor andotify your instructor to add or drop this course. If you are considering a withdrawal, review
the following ASU policies:

Withdrawal from Classes (

Medical/Compassionate Withdrawal (
Grade of Incomplete (

Grade Appeals
Grade disputes must first be addressed by discussing the situation with the instructor. If the dispute is not
resolved with the instructor, the student may appeal to the department chair per the University Policy for
Student Appeal Procedures on Grades (
Student Conduct and Academic Integrity
ASU expects and requires its students to act with honesty, integrity, and respect. Required behavior
standards are listed in the Student Code of Conduct and Student Disciplinary Procedures
(, Computer, Internet, and Electronic
Communications policy (, ASU Student Academic
Integrity Policy (, and outlined by the Office of Student Rights
& Responsibilities ( Anyone in violation of these policies is subject to
Students are entitled to receive instruction free from interference by other members of the class
( An instructor may withdraw a student from the
course when the student's behavior disrupts the educational process per Instructor Withdrawal of a
Student for Disruptive Classroom Behavior (
Appropriate online behavior (also known as netiquette) is defined by the instructor and includes keeping
course discussion posts focused on the assigned topics. Students must maintain a cordial atmosphere and
use tact in expressing differences of opinion. Inappropriate discussion board posts may be deleted by the
instructor. The Office of Student Rights and Responsibilities accepts incident
reports ( from students, faculty, staff, or other persons who
believe that a student or a student organization may have violated the Student Code of Conduct.
Prohibition of Commercial Note Taking Services
In accordance with ACD 304-06 Commercial Note Taking Services
(, written permission must be secured from the
official instructor of the class in order to sell the instructor's oral communication in the form of notes.
Notes must have the note takers name as well as the instructor's name, the course number, and the date.
Course Evaluation
Students are expected to complete the course evaluation. The feedback provides valuable information to
the instructor and the college and is used to improve student learning. Students are notified when the
online evaluation form is available.
Syllabus Disclaimer
The syllabus is a statement of intent and serves as an implicit agreement between the instructor and the
student. Every effort will be made to avoid changing the course schedule but the possibility exists that
unforeseen events will make syllabus changes necessary. Please remember to check your ASU email and
the course site often.
Accessibility Statement
Disability Accommodations: Qualified students with disabilities who will require disability
accommodations in this class are encouraged to make their requests to me at the beginning of the
semester either during office hours or by appointment. Note: Prior to receiving disability
accommodations, verification of eligibility from the Disability Resource Center (DRC) is required.

Disability information is confidential.

Establishing Eligibility for Disability Accommodations: Students who feel they will need disability
accommodations in this class but have not registered with the Disability Resource Center (DRC) should
contact DRC immediately. Students should contact the Disability Resource Center, campus-specific
location and contact information ( can be found on the DRC website.
DRC offices are open 8 a.m. to 5 p.m. Monday Friday. Check the DRC website (
for eligibility and documentation policies.
DRC Phone: (480) 965-1234
DRC FAX: (480) 965-0441
Technical Requirements & Support
Computer Requirements
This course requires Internet access and the following:
A web browser.
Adobe Acrobat Reader (
Adobe Flash Player (
WEKA software
Computer Skills Requirements
It is expected that you will be able to do at least the following tasks on a computer:
Use the Blackboard Learning Management System (see for assistance)
Using ASU email
Creating and submitting files in commonly used word processing program formats (specifically
Microsoft Word)
Copying and pasting text
Downloading and installing software
Using spreadsheet programs (specifically Microsoft Excel)
Using presentation and graphic programs
Technical Support
This course uses Blackboard to deliver course content. It can be accessed through MyASU
at or the Blackboard home page at
To monitor the status of campus networks and services, visit the System Health Portal
at or via Twitter by following @ASUOutages.
To contact the help desk you have two options:
Website: assessed through the MyASU Service Center at
Chat: assessed through the MyASU Service Center at
Call toll-free at 1-855-278-5080
Title IX
Title IX is a federal law that provides that no person be excluded on the basis of sex from participation in,
be denied benefits of, or be subjected to discrimination under any education program or activity. Both
Title IX and university policy make clear that sexual violence and harassment based on sex is prohibited.
An individual who believes they have been subjected to sexual violence or harassed on the basis of sex
can seek support, including counseling and academic support, from the university. If you or someone you
know has been harassed on the basis of sex or sexually assaulted, you can find information and resources

Course Outline and Schedule for Fall 2016

Tan, et al.
Introduction to data mining, basic notation, data
Chapter 1
size, computer software

Tan, et al.


Details for supervised and unsupervised learning

Chapter 2-3

2-6, 2-16, 2-23


Classification and decision trees, modify for

regression, surrogates

Chapter 4

4-2, 4-3, 4-6, 4-8


Evaluating performanceover and under fitting


Classifiers: rule-based, nearest neighbors


Classifiers: nave Bayes

Midterm Exam 9/22


Bayesian networks


Neural networks


Neural networks
No class Oct 11, Fall Break

Chapter 5.4


Support vector machines

Chapter 5.5


Ensemble methods, bagging, boosting

Chapter 5.6-5.8


Association analysis
Midterm Exam 11/03

Chapter 6.16.3, 6.7-6.10

6-2, 6-12, 6-14, 6-16


Association analysisadditional topics


Cluster analysis hierarchical, K-means

Project due 11/19

Chapter 8

8-6, 8-10, 8-11, 8-16, 819


Cluster analysisalternatives
No Class 11/24 Thanksgiving Holiday

Chapter 9.1-9.3 9-3, 9-4


Cluster evaluation, last class 12/02

Final exam, Thursday, 12/8, 2:30 4:20 PM



Chapter 5.1-5.3 5-1, 5-3, 5-4, 5-7

5-12, 5-17, 5-14, 518(a),(b), 5-23

Exercises provided

Exercises provided