Course Form
I. Summary of Proposed Changes
Dept / Program
Computer Science
Course Title
Pattern Recognition
Prefix and Course #
CSCI 448/548
Short Title (max. 26 characters incl. spaces)
Summarize the change(s) proposed
New course
II. Endorsement/Approvals
Complete the form and obtain signatures before submitting to Faculty Senate Office
Please type / print name Signature
Requestor:
Douglas W Raiford
Phone/ email :
243-5605
Douglas.raiford@umontana.com
Program Chair/Director:
Yolanda Reimer
Other affected programs
Dean:
Date
Chris Comer
Are other departments/programs affected by this
Please obtain signature(s) from the
modification because of
Chair/Director of any such department/ program
(a) required courses incl. prerequisites or corequisites,
(above) before submission
(b) perceived overlap in content areas
(c) cross-listing of coursework
III: To Add a New Course Syllabus and assessment information is required (paste syllabus into
section V or attach). Course should have internal coherence and clear focus.
Common Course Numbering Review (Department Chair Must Initial):
YES
NO
Does an equivalent course exist elsewhere in the MUS? Check all relevant disciplines if
X
course is interdisciplinary. (http://mus.edu/transfer/CCN/ccn_default.asp)
If YES: Do the proposed abbreviation, number, title and credits align with existing course(s)? Please indicate
equivalent course/campus. 
If NO: Course may be unique, but is subject to common course review. Be sure to include learning outcomes
on syllabus or paste below. The course number may be changed at the system level.
Proposed course number: CSCI U 448 (and G 548)
Exact entry to appear in the next catalog (Specify course abbreviation, level, number, title, credits,
repeatability (if applicable), frequency of offering, prerequisites, and a brief description.) 
U 448 Pattern Recognition 3 cr. Offered intermittently. Prereq., CSCI 232 (CS 241) or consent of instr. Coconvenes with CSCI 548. Introduction to the framework of unsupervised learning techniques such as
clustering (agglomerative, fuzzy, graph theory based, etc.), multivariate analysis approaches (PCA, MDS,
LDA, etc.), image analysis (edge detection, etc.), as well as feature selection and generation. Emphasis will
be on the underlying algorithms and their implementation. Credit not allowed for both CSCI 448 and CSCI
548.
Justification: How does the course fit with the existing curriculum? Why is it needed?
We currently have a course in the CS curriculum called Machine Learning that covers supervised learning
techniques utilized in the field of artificial intelligence. This proposed course will make a perfect complement
to this course as it will cover unsupervised learning techniques. These techniques are extremely useful when
performing exploratory data analysis on large datasets. This is a skill more and more in demand as data
resulting from experimental trials continues to grow at increasing rates.
Are there curricular adjustments to accommodate teaching this course?
No
Complete for UG courses (UG courses should be assigned a 400 number).
Describe graduate increment - see procedure 301.30
http://umt.edu/facultysenate/committees/grad_council/procedures/default.aspx
Complete for Co-convened courses
Companion course number, title, and description (include syllabus of companion course in section V)
See procedure 301.20 http://umt.edu/facultysenate/committees/grad_council/procedures/default.aspx.
G 548 Pattern Recognition 3 cr. Offered intermittently. Prereq., CSCI 232 (CS 241) or consent of instr.
Co-convenes with CSCI 448. Introduction to the framework of unsupervised learning techniques such as
clustering (agglomerative, fuzzy, graph theory based, etc.), multivariate analysis approaches (PCA, MDS,
LDA, etc.), image analysis (edge detection, etc.), as well as feature selection and generation. Techniques in
exploratory data analysis when faced with large, multivariate datasets. Opportunities at implementation of
some algorithmic approaches as well as use of preexisting tools such as the R-project statistics package.
Emphasis will be on the underlying algorithms and their implementation. Credit not allowed for both CSCI
448 and CSCI 548.
New fees and changes to existing fees are only approved once each biennium by the
YES
NO
Board of Regents. The coordination of fee submission is administered by
Administration and Finance. Fees may be requested only for courses meeting specific
conditions according to Policy 940.12.1 http://mus.edu/borpol/bor900/940-12-1.pdf .
Please indicate whether this course will be considered for a fee.
If YES, what is the proposed amount of the fee?
Justification:
X
IV. To Delete or Change an Existing Course – check X all that apply
Deletion
Title
Course Number Change
From:
Level U, UG, G
Co-convened
To:
Description Change
Change in Credits
From:
To:
Prerequisites
1. Current course information at it appears in catalog
(http://www.umt.edu/catalog) 
From:
To:
Repeatability
Cross Listing
(primary
program initiates
form)
Is there a fee associated with the course?
2. Full and exact entry (as proposed) 
3. If cross-listed course: secondary program & course
number
4. If co-convened course: companion course number, title, and description
(include syllabus of companion course in section V) See procedure 301.20
http://umt.edu/facultysenate/committees/grad_council/procedures/default.aspx.
5. Is this a course with MUS Common Course Numbering?
http://mus.edu/transfer/CCN/ccn_default.asp
If yes, please explain below whether this change will eliminate the course’s common course
status.
6. Graduate increment if level of course is changed to
UG. Reference procedure 301.30:
http://umt.edu/facultysenate/committees/
grad_council/procedures/default.aspx
YES NO
Have you reviewed the graduate increment
guidelines? Please check (X) space
provided.
(syllabus required in section V)
7. Other programs affected by the change
8. Justification for proposed change
V. Syllabus/Assessment Information
Required for new courses and course change from U to UG. Paste syllabus in field below or attach and send
digital copy with form.
Syllabus for Pattern Recognition
Pattern Recognition
CSCI 448/548
Spring 2013
Course Machine Learning: CSCI 448/548
Time TBD
Room TBD
Instructor Dr. Douglas W. Raiford
Office Social Science 412
Phone 406-243-5605
Email Douglas.Raiford@umontana.edu
Office Hours TBD
Instructor Website http://www.cs.umt.edu/~dougr/
Pattern Recognition
Sergios Theodoridis and Konstantinos Koutroumbas
Author’s Website
Textbook
Publisher:
Academic Press; 3 edition (March 10, 2006)
Language:
English
ISBN-10: 0123695317
ISBN-13: 978-0123695314
Grading
Component
Undergraduate
Graduage
Homework
10%
10%
Quizzes
15%
10%
Two Exams & Final 35%
35%
Projects
40%
35%
Grad Student
Project
NA
10%
90 - 100 ................................ A
87 - 89 .................................. B+
80 - 86 .................................. B
77 - 79 .................................. C+
Scale:
70 - 76 .................................. C
67 - 69 .................................. D+
60 - 66 .................................. D
00 - 59 .................................. F


Course Objectives


Prerequisites
Instill in the students an understanding of where Pattern
Recognition sits in the hierarchy of artificial intelligence
and soft computing techniques
Develop expertise in various unsupervised learning
algorithms such as clustering techniques (agglomerative,
fuzzy, graph theory based, etc.), multivariate analysis
approaches (PCA, MDS, LDA, etc.), image analysis (edge
detection, etc.), as well as feature selection and
generation.
Provide a tool-kit of pattern recognition problem solving
approaches that the students can take with them for use in
future research or other programmatic endeavors
Provide the student with the ability to apply these
techniques in exploratory data analysis.
CSCI 232 Data Structures and Algorithms (or consent of
instr.)
1. Introduction/definition of pattern recognition
2. Overview of supervised vs. unsupervised learning
techniques
3. Clustering overview (categories, proximity measures, etc.)
4. Sequential clustering approaches
5. Hierarchical clustering approaches
6. Cost function optimization clustering approaches
Topics 7. Probabilistic clustering approaches
8. Cluster validity metrics
9. Multivariate analysis techniques
10. Principal Components Analysis
11. Multidimensional scaling
12. Linear Discriminant Analysis
13. Image analysis techniques
14. Feature selection and generation
Graduate students (attending CSCI 548) will have the added
responsibility of completing a graduate level project. The
motivation for this activity is to provide additional experience
in the field of pattern recognition in such a way as to promote
the students ability to synthesize new approaches based upon
the concepts encountered within the course.
The graduate level project will have three components to it: 1)
Graduate Increment a data component 2) a code component 3) an analysis
component.
Around the 5th week of the course you must decide upon what
you will do for your project, and submit a proposal. The ideal
project is one where you already have some data (and/or a
computational goal) that is part of your graduate research. In
this way you will be applying pattern recognition techniques
to data and analysis that will assist you in your thesis work.
Policies
The students attending this class are at the graduate level or
are experienced undergrads, and are expected to perform at
these levels. Assignment submissions should be on time.
Failure to do so is an indication of poor time management and
Late assignments
lack of effort. While each case will be treated independently,
and the situation will be discussed with the student, a penalty
of roughly a letter grade per day late will be imposed on the
assignment.
Ethics in academic activities are important at the University
of Montana. We wish to graduate students who are
responsible, hard working, dependable, and who exhibit
integrity and independence of thought.
The assignments and exams given in this course are designed
to reinforce your learning and measure your understanding the
topics covered in class. As such, the work you turn-in should
be your own, and no one else’s.
Academic Dishonesty
and the Honor Code Overly similar work will be considered to be the result of
copying. If you collaborate with another person for a graded
assignment as in the example activities noted above, all
parties involved will receive a zero for that assignment. If
there are further assignments in which you have collaborated,
the matter will be turned over to the Dean of Academic
Affairs for possible university imposed sanction. It is,
therefore, imperative that if you need help on your
assignments that you contact your instructor or TA and
NOT someone else. The official University policies can be
found in the Student Conduct Code.
The Department of Computer Science is committed to equal
opportunity in education for all students, including those with
documented physical disabilities or documented learning
disabilities. University policy states that it is the responsibility
of students with documented disabilities to contact instructor
DURING THE FIRST WEEK OF THE SEMESTER to
discuss appropriate accommodations to ensure equity in
Accommodations
grading, classroom experiences, and outside assignments.
The instructor will meet with the student and the staff of the
Disability Services for Students (DSS) to make
accommodations. Please contact Jim Marks in DSS
(243.2373, Lommasson Center 154) for more information.
Religiously observant students wishing to be absent on
holidays that require missing class should notify their
Religious observances professors in writing at the beginning of the semester, and
should discuss with them, in advance, acceptable ways of
making up any work missed because of the absence.
Students participating in an officially sanctioned, scheduled
University extracurricular activity will be given the
opportunity to make up class assignments or other graded
Excused Absences for
assignments missed as a result of their participation. It is the
University
responsibility of the student to make arrangements with the
Extracurricular Activities
instructor prior to any missed scheduled examination or other
missed assignment for making up the work.
VI Department Summary (Required if several forms are submitted) In a separate document list course
number, title, and proposed change for all proposals.
VII Copies and Electronic Submission. After approval, submit original, one copy, summary of
proposals and electronic file to the Faculty Senate Office, UH 221, camie.foos@mso.umt.edu.
Revised 5-4-11