UCC/UGC/ECCC Proposal for New Course Please attach proposed Syllabus in approved university format. 1. Course subject and number: EE 543 2. Units: See upper and lower division undergraduate course definitions. 3. College: CEFNS 4. Academic Unit: 3 Electrical Engineering and Computer Science 5. Student Learning Outcomes of the new course. (Resources & Examples for Developing Course Learning Outcomes) At the completion of this course, students will understand: the probabilistic nature of real signals the concept of features and feature space how classifiers use discriminant functions to partition feature space Bayes rule and the foundations of Bayesian decision theory correlation, normalized cross-correlation Bayesian classification maximum likelihood classification K-mean clustering and classification linear and simple nonlinear discriminants relaxation techniques minimum squared error techniques the basics of neural networks the basics of hidden Markov models the basics of machine learning and classification how to read and gain understanding from current pattern recognition literature Students will also develop their abilities to read and gain understanding from current pattern recognition literature and develop deeper understanding of the above topics in order to help students in the undergraduate section with their in class exercises, provide them with constructive feedback on their term papers, and learn how to, and learn how to teach them via a tutorial and software demonstration. 6. Justification for new course, including how the course contributes to degree program outcomes, or other university requirements / student learning outcomes. (Resources, Examples & Tools for Developing Effective Program Student Learning Outcomes). Pattern Recognition deals with the mathematics, algorithms, and practical implementation details of detecting important information in the presence of noise and other sources of confusion. It has wide relevance in such diverse specialty areas as signal processing, Effective Fall 2012 communications, computer vision, and statistical analysis. Applications include face recognition, speech recognition, gesture recognition, financial analysis, data mining, and robotics. Pattern Recognition theory and techniques lie at the heart of these and many other applications that our students will encounter in their careers. 7. Effective BEGINNING of what term and year? See effective dates calendar. Fall 2013 8. Long course title: PATTERN RECOGNITION (max 100 characters including spaces) 9. Short course title: PATTERN RECOGNITION (max. 30 characters including spaces) 10. Catalog course description (max. 60 words, excluding requisites): Survey of techniques for identifying patterns present in noisy signal and image data. Includes classifiers, discriminant functions, Bayesian decision theory, maximum likelihood, K-means, relaxation, neural networks, and machine learning. Co convenes with EE 443. Letter grade only. 11. Will this course be part of any plan (major, minor or certificate) or sub plan (emphasis)? Yes If yes, include the appropriate plan proposal. No 12. Does this course duplicate content of existing courses? Yes No If yes, list the courses with duplicate material. If the duplication is greater than 20%, explain why NAU should establish this course. This course does not significantly duplicate material in existing courses, It does co-convene with the proposed EE 443 Pattern Recognition course. Additionally, it will cover in much greater depth the basic material of pattern recognition and classification also covered in the proposed EE 444/544 Computer Vision course. This foundational overlap of approximately 15% is necessary to ensure success in each class. 13. Will this course impact any other academic unit’s enrollment or plan(s)? If yes, include a letter of response from each impacted academic unit. 14. Grading option: Letter grade Pass/Fail Yes No Both 15. Co-convened with: EE 443 14a. UGC approval date*: (For example: ESE 450 and ESE 550) See co-convening policy. *Must be approved by UGC before UCC submission, and both course syllabi must be presented. 16. Cross-listed with: (For example: ES 450 and DIS 450) See cross listing policy. Effective Fall 2012 Please submit a single cross-listed syllabus that will be used for all cross-listed courses. 17. May course be repeated for additional units? 16a. If yes, maximum units allowed? 16b. If yes, may course be repeated for additional units in the same term? Yes No Yes No EE 325, EE348 with grades of C or 18. Prerequisites: better. If prerequisites, include the rationale for the prerequisites. The prerequisite of EE 348 provides the necessary background in linear systems, discrete-time signals, and convolution to form a good basis for exploring the pattern recognition topic. EE 348 and EE 325 also provide an important level of mathematical maturity that is vital to understanding the material in this course. EE 348 requires a background and facility with computer programming, a skill that is important in this course. 19. Co requisites: If co requisites, include the rationale for the co requisites. 20. Does this course include combined lecture and lab components? Yes If yes, include the units specific to each component in the course description above. 21. Names of the current faculty qualified to teach this course: No Dr. Phillip Mlsna, David Scott Answer 22-23 for UCC/ECCC only: 22. Is this course being proposed for Liberal Studies designation? If yes, include a Liberal Studies proposal and syllabus with this proposal. Yes No 23. Is this course being proposed for Diversity designation? If yes, include a Diversity proposal and syllabus with this proposal. Yes No FLAGSTAFF MOUNTAIN CAMPUS Scott Galland Reviewed by Curriculum Process Associate 03/21/2013 Date Approvals: 2-14-2013 Department Chair/Unit Head (if appropriate) Effective Fall 2012 Date Chair of college curriculum committee Date Dean of college Date For Committee use only: UCC/UGC Approval Date Approved as submitted: Yes No Approved as modified: Yes No EXTENDED CAMPUSES Reviewed by Curriculum Process Associate Date Approvals: Academic Unit Head Date Division Curriculum Committee (Yuma, Yavapai, or Personal Learning) Date Division Administrator in Extended Campuses (Yuma, Yavapai, or Personal Learning) Date Faculty Chair of Extended Campuses Curriculum Committee (Yuma, Yavapai, or Personal Learning) Date Chief Academic Officer; Extended Offices (or Designee) Date Approved as submitted: Yes No Approved as modified: Yes No Effective Fall 2012 College of Engineering, Forestry & Natural Sciences Department of Electrical Engineering & Computer Science COURSE SYLLABUS: EE 543 PATTERN RECOGNITION General Information: Sequence number: Class times: 3.0 credit hours. There is no laboratory component to this course. Instructor: Dr. Phillip Mlsna, Associate Professor of Electrical Engineering Office: Engineering room 257, 523-2112, phillip.mlsna@nau.edu Office hours as posted (office door and BlackboardLearn) Official course webpages are on BlackboardLearn: http://bblearn.nau.edu Course Prerequisites: EE 348 (Signals and Systems) with grade C or better. EE 325 (Engineering Analysis II) with grade C or better. You are also expected to have good programming skills in both Matlab and C. Course Description (from catalog) » Survey of techniques for identifying patterns present in noisy signal and image data. Includes classifiers, discriminant functions, Bayesian decision theory, maximum likelihood, K-means, relaxation, neural networks, and machine learning. Co-convenes with EE 443. Letter grade only. Student Learning Expectations/Outcomes for this Course : At the completion of this course, students will understand: the probabilistic nature of real signals the concept of features and feature space how classifiers use discriminant functions to partition feature space Bayes rule and the foundations of Bayesian decision theory correlation, normalized cross-correlation Bayesian classification maximum likelihood classification K-mean clustering and classification linear and simple nonlinear discriminants relaxation techniques minimum squared error techniques the basics of neural networks the basics of hidden Markov models the basics of machine learning and classification how to read and gain understanding from current pattern recognition literature Students will also develop their abilities to read and gain understanding from current pattern recognition literature and develop deeper understanding of the above topics in order to help students in the undergraduate section with their in class exercises, provide them with Effective Fall 2012 constructive feedback on their term papers, and learn how to, and learn how to teach them via a tutorial and software demonstration. Course Structure/Approach: We will be following the textbook rather closely most of the time, with the topic order as shown in the “Course Outline” section below. The format will largely be lecture and discussion. The textbook readings are especially important. There will often be important material in the text that we will not have time to cover in class. Required Materials: Duda and Hart, “Pattern Classification”, 2001, Wiley Interscience, ISBN 0-471-05669-3. Current journal articles or recent conference papers to be selected by the students and approved by the instructor for extra homework for those taking the graduate section. Recommended optional materials/references: Theodoridis and Koutroumbas, “Pattern Recognition,” 4th ed., 2009, Academic Press, ISBN 978-159749-272-0. Course Outline Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Week 8 Week 9 Week 10 Week 10 Week 11 Week 12 Week 13 Week 14 Week 15 Week 16 Overview Chapter 1 Bayesian Decision Theory Chapter 2 Maximum Likelihood Estimation Chapter 3 Bayesian Parameter Estimation Chapter 3 Nonparametric Techniques Chapter 4 Exam 1 Linear Discriminant Functions Chapter 5 Multilayer Neural Networks Chapter 6 Training Neural Networks Chapter 6 Stochastic Methods Chapter 7 Nonmetric Methods Chapter 8 Exam 2 Intro to Machine Learning Chapter 9 Unsupervised Learning and Clustering Chapter 10 Hidden Markov Models, project presentations lectures Project & term paper presentations, review lectures Final Exam Assessment of Student Learning Outcomes: Assessment will be based on two mid-term exams, homework, participation, a term project, and a comprehensive final exam. Three article reports will require the selection, reading, and comprehension of materials from the recent research literature in computer vision. Students in EE 543 will also help guide the EE 443 students with in-class exercises and with their EE 443 term papers. EE 543 students will each review several EE 443 draft term papers and provide constructive feedback. Term Project: For those in the graduate section, a semester project involving a deeper investigation into a relevant topic and demonstration of software is required. Students will work individually on the term project. The student will present a background tutorial to the class along with his/her project results and a Effective Fall 2012 demonstration. In this way, graduate students will provide some educational benefit to the undergraduates in EE 443. Grading System and Assessment Timing: Exam 1 125 points Exam 2 125 points Final exam 200 points Homework 120 points Term Project 100 points PR Journal Article Reports 30 points Participation 25 points Leadership 25 points Total 750 points approximately week 6 (25 points of unique or additional problems for graduate section) approximately week 11(25 points of unique or additional problems for graduate section) comprehensive (50 points of unique or additional problems for graduate section) approximately once per week (20 points of unique or additional assignments for graduate section) semester project instead of a term paper (advanced material and classroom teaching/demonstration for 25 more points for those in the graduate section) three written reports for those in the graduate section attendance and active classroom participation leadership activities for those in the graduate section Course Policies: Late Work Assignments are due when specified and can be submitted on BBLearn (preferred) or on paper at the beginning of the class period. Late work will be accepted electronically only (on BBLearn, not by e-mailing the professor!) up to 24 hours late for a 20% penalty, and not accepted after 24 hours late. Retests and Makeup Tests No makeup exams will be given except by prior arrangement in exceptional or emergency situations at the discretion of the instructor. Please contact me immediately if such a situation arises. (Procrastination is not an emergency.) Attendance Attendance is required and will be recorded on a random basis. Attendance data will be included in the participation portion of your grade. Academic Dishonesty Cheating and plagiarism are strictly prohibited. Incidents of cheating or plagiarism are treated quite seriously. The NAU policy on academic dishonesty in Appendix G of the current Student Handbook applies. All work you submit for grading must be your own. http://home.nau.edu/studentlife/handbook/appendix_g.asp You are encouraged to discuss the intellectual aspects of homework assignments with other class participants. However, each student is responsible for formulating solutions in his or her own words. University policies: Safe Working and Learning Environment Students with Disabilities Institutional Review Board Academic Integrity Academic Contact Hour Sensitive Course Material Effective Fall 2012 See the following document for these policy statements: http://www4.nau.edu/avpaa/UCCPolicy/plcystmt.html. Effective Fall 2012