1. South Dakota School of Mines and Technology
CSC 448/548 - Machine Learning
Fall 2007
Instructor: Dr. Manuel L. Penaloza
Office: M-312
Office hours: Monday 1-2 p.m., Tuesday 2-4 p.m., Thursday 10-noon., or by appointment.
Office Phone: 394-6077
e-mail: Manuel.Penaloza@sdsmt.edu
URL: http://sdmines.sdsmt.edu/sdsmt/directory/courses/2007fa/csc448/548M001
Class Schedule
Lecture: MWF 10:00 – 10:50 am [MM 213]
Current Catalog Description
A systematic study of the theory and algorithms that constitute machine learning. It covers
learning based on examples including genetic algorithms, case-based reasoning, decision
trees, and Bayesian methods. Students enrolled in CSC 548 will be held to a higher
standard than those enrolled in CSC 448.
Prerequisite: CSC 300.
Recommended Textbook
Introduction to Machine Learning by Ethem Alpaydin. The MIT Press, 2004. ISBN:
0-262-01211-1.
References
Tom Mitchell. Machine Learning, WCB/McGraw-Hill, 1997.
Janet Kolodner. Case-Based Reasoning. Morgan Kaufmann Publishers, Inc., 1993.
Ian Watson. Applying Case-based reasoning. Morgan Kaufmann Publishers, 1997.
David Goldberg, Genetic Algorithms, in search, Optimization and Machine Learning.
Addison Wesley, 1989.
Course Goals
Computer programs that can learn by experience would have an impact not only on the cost
reduction of the software, but also in the reduction of bugs that a programmer usually introduce
when he or she write computer programs. In this course we will cover the foundations of machine
learning, and a variety of machine learning algorithms. Students will have the opportunity to have
hands on experience with several of these algorithms by implementing assignments or projects
using a tool that consists of several machine learning algorithms called Weka, which is freely
available for download from http://www.cs.waikato.ac.nz/ml/weka/. Several sources of lectures
notes will be used for this class. Links to their sources or electronic copies will appear in the course
website.
Topics
Introduction to Machine Learning
A variety of types of learning: Supervised, unsupervised, reinforcement, concept, instance-
based, Bayesian, and others.
Clustering

2. Regression and classification
Artificial neural networks
Dimensionality reduction
Feature extraction
Model evaluation
Case-based reasoning
Evolutionary algorithms: Biology concepts, genetic algorithms, and genetic programming.
Decision trees
Course Outcome
Upon completion of this course, students will be able to:
1. Describe the goals of machine learning
2. Describe the learning process
3. Describe the components and classes of evolutionary algorithms
4. Learn different types of machine learning algorithms
5. Construct fitness functions, selection and genetic operators for a given problem
6. Learn implementation details of several machine learning algorithms that are used in fields
such as data mining, pattern recognition, and others
7. Learn how to reduce and extract features from a dataset
8. Discover patterns or features of datasets
9. Learn how to select and evaluate a machine learning model
10. Apply machine learning algorithms to datasets
11. Generate experimental results with a machine learning programming tool
Grading Criteria
Grading will be based on students combined performance in homework assignments,
student participation in class activities, a final project, a mid-term exam, and a cumulative
final exam. Some of the assignments require the use of the Weka tool. For the project,
students must select a machine learning topic, read at least three papers related to the
selected topic, implement or find the implementations of at least two different algorithms,
and find or generate datasets to compare these algorithms by running their
implementations. You must write a report of your work. You must include copies of the
papers. Graduate students in addition to the report, must do a presentation of the work to
the class. The last week of class will be assigned for these presentations. There exist
several journals on machine learning. Try to find them online, or get them through the
school library. Teams of up to two students are allowed per project. Instructions for the
project and deadlines for the report and presentation will be given before midterm. The
tentative schedule for the exams is mid-tem exam: 10/10, and the final exam: 2-3:50 p.m.
on Monday December 17, 2007.
Midterm exam ......… 15%
Final exam …….……. 25%
Class activities ……… 5%
Assignments ………... 30%
Project ……………..... 25%
The grading scale is:
A : 90-up B : 80.00-89.99 C : 70.00-79.99
D : 60.00-69.99 F : 0.00-59.99
Attendance

3. Attendance is required for all courses at SDSM&T. This course will include activities in addition
to lectures. You are responsible for the lecture material as well as the assigned readings in the
textbook.
Special Requirements
Students with special needs or requiring special accommodations should contact the instructor,
and/or the campus ADA coordinator, Ms. Jolie McCoy, at 394-1924 at the earliest opportunity.
Electronic Device Policy
Please turn off your cell phone before class starts. No text messaging in class. No
headphones. If you wish to use a laptop in this class for purposes of note taking, that’s
great; however, you will be required to download DyKnow software and then join
ENGL350 to activate. Any attempt to circumvent the DyKnow monitoring system will be
considered a form of cheating and a breach of academic integrity. Note that according to
“Policy Governing Academic Integrity” in the SDSM&T Undergraduate Catalog, the
instructor of record for this course has discretion of how acts of academic dishonesty are
penalized, subject to the appeal process, and that “Penalties may range from requiring the
student to repeat the work in question to failure in the course” (72-73). No other use of
any other electronic/computer media is allowed during class time.
Freedom in Learning Statement
Under Board of Regents and University policy student academic performance may be
evaluated solely on an academic basis, not on opinions or conduct in matters unrelated to
academic standards. Students should be free to take reasoned exception to the data or
views offered in any course of study and to reserve judgment about matters of opinion, but
they are responsible for learning the content of any course of study for which they are
enrolled. Students who believe that an academic evaluation reflects prejudiced or
capricious consideration of student opinions or conduct unrelated to academic standards
should contact the dean of the college which offers the class to initiate a review of the
evaluation.
NOTES and POLICIES:
1) If you are having troubles with an assignment come see the instructor immediately. The
instructor is aware there is not enough time in the classroom for everyone.
2) All work must be handed in at the beginning of class on the due date. The instructor will not
accept any work of a student who is not present in class.
3) Late assignments are not accepted. Any type of assignment not turned in or exams not taken
count as ZEROS. There are NO MAKEUP assignments or programming projects.
4) MAKEUP exams will be given only if you contact the instructor or the Mathematics and
Computer Office (394-2471) BEFORE the exam is scheduled to start and provide a legitimate
reason.
5) Some course material, such as assignments, solutions will be published at the course’s web
site.
6) The work to be handed in for this class must be an individual effort unless the instructor has
explicitly stated otherwise. The instructor expects every student to produce his/her own
solution and work. Coping someone else’s code or work is not acceptable. The minimum
penalty for any violation to this policy will be a zero to the assignment, project, or exam and
one lower final letter grade.
7) Each completed project should be well documented. The assignment sheet will indicate what
is expected for documentation.
The school has computer, software, network, and academic conduct guidelines and policies. Please
make sure you are well familiar with them and follow them.