Download Attachments - College of Engineering and Computer Science

New Course Proposal – Page 1/8
NEW COURSE PROPOSAL
College:
Engineering and Computer Science ]
[ Department: [ Computer Science ]
Note: Use this form to request a single course that can be offered independently of any other course, lab or activity.
1. Course information for Catalog Entry
Subject Abbreviation and Number: [ COMP 541 ]
Course Title: [ DATA MINING ]
Units: [ 3 ] units
Course Prerequisites: [ COMP 380/L ] (if any)
Course Corequisites: [
] (if any)
Recommended Preparatory Courses: [
] (if any)
2. Course Description for Printed Catalog: Notes: If grading is NC/CR only, please state in course description.
If a course
numbered less than 500 is available for graduate credit, please state “Available for graduate credit in the catalog description.”
[ A study of the concepts, principles, techniques, and applications of data mining. Topics include
data preprocessing, the ChiMerge algorithm, data warehousing, OLAP technology, the Apriori
algorithm for mining frequent patterns, classification methods (such as decision tree induction,
Bayesian classification, neural networks, support vector machines, genetic algorithms), clustering
methods (such as k-means algorithm, hierarchical clustering methods, self-organizing feature map),
and data mining applications (such as Web, finance, telecommunication, biology, medicine, science,
and engineering). Privacy protection and information security in data mining are also discussed. ]
3. Date of Proposed Implementation: (Semester/Year): [ Fall ] / [ 2011 ] Comments
4. Course Level
[ ]Undergraduate Only
[
]Graduate Only
[
]Graduate/Undergraduate
5. Course Abbreviation “Short title” (maximum of 17 characters and spaces)
Short Title: [ D•A•T•A• •M•I•N•I•N•G• • • • • • ]
6. Basis of Grading:
[ ]Credit/No Credit Only
[
]Letter Grade Only
[
]CR/NC or Letter Grade
7. Number of times a course may be taken:
[
] May be taken for credit for a total of [1] times, or for a maximum of [3] units
[
] Multiple enrollments are allowed within a semester
8. C-Classification: (e.g., Lecture-discussion (C-4).)
[ 3 ] units @ [C-5] [ ]
9. Replaces Current Experimental Course?
[
] YES
[
] NO
Replaces Course Number/Suffix:[ COMP 595DM ]
Previously offered [ 3 ] times.
NC – 9/29/05
New Course Proposal – Page 2/8
10. Proposed Course Uses: (Check all that apply)
[
]Own Program:
[
]Major
[
]Minor
[
]Masters
[
] Requirement or Elective in another Program
[
] General Elective
[
] General Education, Section [
]
[
] Meets GE Information Competence (IC) Requirement
[
] Meets GE Writing Intensive (WI) Requirement
[
] Community Service Learning (CS)
[
] Cross-listed with: (List courses) [
]
[
]Credential
[
]Other
11. Justification for Request: Course use in program, level, use in General Education, Credential, or other. Include
information on overlap/duplication of courses within and outside of department or program. (Attach)
12. Estimate of Impact on Resources within the Department, for other Departments and the
University. (Attach)
(See Resource List)
13. Course Outline and Syllabus (Attach) Include methods of evaluation, suggested texts, and selected bibliography.
Describe the difference in expectations of graduates and undergraduates for all 400 level courses that are offered to both.
14. Indicate which of the PROGRAM’S measurable Student Learning Outcomes are addressed in
this course. (Attach)
15. Assessment of COURSE objectives (Attach)
A. Identify each of the course objectives and describe how the student performance will be
assessed
(For numbers 14 and 15, see Course Alignment Matrix and the Course Objectives Chart)
16. If this is a General Education course, indicate how the General Education Measurable Student
Learning Outcomes (from the appropriate section) are addressed in this course. (Attach)
17. Methods of Assessment for Measurable Student Learning Outcomes (Attach)
A. Assessment tools
B. Describe the procedure dept/program will use to ensure the faculty teaching the course will be
involved in the assessment process (refer to the university’s policy on assessment.)
18. Record of Consultation: (Normally all consultation should be with a department chair or program coordinator. ) If
more space is needed attach statement and supporting memoranda.
Department Chair/ Program
Concur
Date:
Dept/College:
Coordinator
(Y/N)
[ 2/5/2010 ]
[ Computer Science ]
[ Dept vote; Steven Stepanek ]
[Y]
[ 3/12/2010 ]
[ ME ]
[ Hamid Johari ]
[Y]
[ 3/12/2010 ]
[ MSEM ]
[ Behzad Bavarian ]
[Y]
[ 3/12/2010 ]
[ ECE ]
[ Ali Amini ]
[Y]
[ 3/12/2010 ]
[ CEAM ]
[ Steve Gadomski ]
[Y]
[
]
[
]
[
]
[Y]
Consultation with the Oviatt Library is needed to ensure the availability of appropriate resources to
support proposed course curriculum.
Collection Development Coordinator, Mary Woodley
Date
Please send an email to: [email protected]
[
NC – 9/29/05
New Course Proposal – Page 2/8
3/8/2010
]
19. Approvals:
Department Chair/Program Coordinator:
Date:
College (Dean or Associate Dean):
Date:
Educational Policies Committee:
Date:
Graduate Studies Committee:
Date:
Provost:
Date:
NC – 9/29/05
[ 3/8/2010 ]
[
]
[
]
[
]
[
]
New Course Proposal – Page 3/8
Attachments
11. Justification for Request
Currently we are facing an era of “data flood” and “information lack.” Addressing that challenge,
industry has recently adapted data mining technologies. In the Information Age, information and
knowledge are essential assets for the various areas such as business, telecommunication, finance, health,
science, engineering, national securities and the like. Data mining discovers information and knowledge
which are hidden in a huge pile of data. For instance, data mining technologies are widely used
efficiently in the world-class companies such as Wal-Mart, Amazon.com, eBay, Citibank, State Farm,
Amgen, to name just a few. By using data mining technologies, these companies are establishing
priceless knowledge assets that can be applicable to their marketing, management, finance, and sales.
In academia, data mining became not only a new and advanced field in computer science, but also a
supporting tool for other disciplines such as biology, chemistry, finance, etc., which deal with large data
sets. For example, thanks to advanced computer technology, a large amount of genetic data has been
generated and accumulated into databases, and biologists are trying to find genetic functions from the
genetic data by using data mining tools. To support this demand, many colleges and universities are
currently providing data mining courses.
Recognizing this trend, I submitted an experimental course proposal for data mining in Fall 2005, and the
proposal was approved. This course was offered in Fall 2006, Spring 2008, and Fall 2009, and the results
turned out to be successful in terms of enrollment and student feedback.
The demand of data mining technologies is expected to remain strong and to continue to grow in the
future. To cope with this demand, we need to replace the current experimental course with a regular
course. Meanwhile, by placing this course at the 500 level, undergraduate students as well as graduate
students can take this course as one of their electives.
12. Estimate of Impact on Resources within the Department, for other Departments and the
University
The impact of this course on the resources within the department are minimal. For the past three years it
has been offered as an experimental course in the fall; as a regular course it would continue to be offered
once per academic year; so the future impact will not be different from its current impact. No resource
impact is anticipated on other departments or the University.
13. Course Outline and Syllabus
Following is the COMP 595DM syllabus used in Fall 2009.
Syllabus
COMP 595DM: Data Mining
Fall 2009
Course Information
- Textbook:
NC – 9/29/05
Reference:
Jiawei Han and Micheline Kamber, Data Mining: Concepts and Techniques,
2nd Edition, Morgan Kaufmann, 2006
Toby Segaran, Programming Collective Intelligence, O’Reilly, 2007
New Course Proposal – Page 4/8
-
Instructor:
Class website
Class number
Lecture:
Office hour:
Midterm exam:
Final exam:
Prerequisite:
George (Taehyung) Wang, [email protected]
http://www.csun.edu/~twang/595DM
18900
JD3510, M 19:00 – 21:45
JD4447, MW 14:45- 15:45, or by appointment
To be announced
To be announced
COMP380/L
Course Description
A study of the concepts, principles, techniques, and applications of data mining. Topics include
data preprocessing, the ChiMerge algorithm, data warehousing, OLAP technology, the Apriori
algorithm for mining frequent patterns, classification methods (such as decision tree induction,
Bayesian classification, neural networks, support vector machines, genetic algorithms), clustering
methods (such as k-means algorithm, hierarchical clustering methods, self-organizing feature
map), and data mining applications (such as Web, finance, telecommunication, biology, medicine,
science, and engineering). Privacy protection and information security in data mining are also
discussed.
Course Objectives
Upon successful completion of the course the student will:
- Be able to understand the concepts, strategies, and methodologies related to the design and
construction of data mining.
- Be able to comprehend several data preprocessing methods including the ChiMerge algorithm.
- Be able to utilize data warehousing and OLAP technologies.
- Be able to apply appropriate mining techniques to extract unexpected patterns and new rules
that are hidden in large databases.
- Be able to understand current and future data mining technologies.
Course Requirements
Attendance
Attendance is a must.
Homework Assignments
Several homework assignments will be given to assess the understanding of class materials.
Homework solutions should be submitted before the lecture starts on the due date. Occasionally
presentation will be asked and counted toward the grade. The evaluation form is available at
http://www.csun.edu/~twang/595DM/HomeworkAssignments/PresentationEvaluationForm.pdf.
Group Project
As one of the course requirements, the success of the group project is necessary. The students
must successfully finish their group projects on time. The topic of the group project should be
approved by Week 3. Peer evaluation will be conducted during the final exam.
Term Paper
The result of a group project will be shown in a term paper. The format of the term paper will be
available the class web site. Term paper judging criteria are available at
http://www.csun.edu/~twang/595DM/TermPaper/JudgingCriteria.pdf. Peer evaluation will be
conducted during the final exam.
NC – 9/29/05
New Course Proposal – Page 5/8
Midterm Exam
There will be one midterm in the middle of the semester. The format of the exam will be
announced in advance. No makeup exam is allowed unless a legal document is provided to the
instructor.
Final Exam
The scope of final exam is comprehensive. The format will be announced before final exam week.
No makeup exam is allowed unless a legal document is provided to the instructor.
Course Policies
Attendance
Without an advanced notice with reasonable excuse, absences affect your final grade. The
instructor will check attendance at the beginning of lectures and labs.
Academic Dishonesty
“Cheating or plagiarism in connection with an academic program at a campus is listed in Section
41301, Title V, California Code of Regulations, as an offense for which a student may be
expelled, suspended, or given a less severe disciplinary sanction.” (http://www.csun.edu/catalog/)
Grading
The points of each grading components, and the grading formula are as follows:
Grading components
Homework assignments
Midterm exam
Term paper
Group project
Final exam
Total
Percentage
25%
15%
15%
15%
30%
100%
Grade
A
AB
C
D
F
Grade cutoff
>= 93.00 %
>= 90.00 %
80 – 89.99%
70 – 79.99%
60 – 69.99%
< 60%
Note that the numbers homework assignments are tentative; the grading formula can be changed
depending on a circumstance and instructor’s decision.
Course Schedule and Outline
Week
Topics
Week 1
Introduction to data mining and applications: Ch. 1 & Ch.11
Week 2
Data type and data storage: Ch. 1; Data preprocessing: Ch. 2. (Descriptive data
summarization, data cleaning, data transformation)
Week 3
Labor day
Week 4
Data reduction, data discretization, and concept hierarchy: Ch. 2
Week 5
Data warehouse: Ch. 3 (Concept, multidimensional data model)
Week 6
Mining association rules: Ch. 5
Week 7
Decision tree: Ch. 6
Week 8
Bayesian network: Ch. 6
Week 9
Neural network: Ch. 6
Week 10
Bayesian network: Ch. 6
Week 11
Support vector machine: Ch. 6
NC – 9/29/05
New Course Proposal – Page 6/8
Week 12
Week 13
Week 14
Week 15
Week 16
Support vector machine: Ch. 6
Genetic algorithm: Ch. 6
k-means algorithm and hierarchical algorithm: Ch. 7
Self Organizing Maps: Ch. 7
Privacy protection and information security in data mining: Ch. 11
*Note that the course schedule and outline can be changed without advance notice.
14. Indicate which of the PROGRAM’S measurable Student Learning Outcomes are addressed in
this course.
As a result of their participation in the program, students will be able to:
- Apply knowledge of computing and mathematics appropriate to the disciplines (SLO a)
- Analyze a problem, and identify and define the computing requirements appropriate to its
solutions (SLO b)
- Design, implement, and evaluate a computer-based system, process, component, or program to
meet desired needs (SLO c)
- Function effectively on teams to accomplish a common goal (SLO d)
- Understand professional, ethical, legal, security, and social issues and responsibilities (SLO e)
- Communicate effectively with a range of audience (SLO f)
- Analyze the local and global impact of computing on individuals, organizations, and society
(SLO g)
- Recognize the need for and an ability to engage in continuing professional development (SLO
h)
- Use current techniques, skills, and tools necessary for computing practices (SLO i)
- Apply mathematical foundations, algorithmic principles, and computer science theory in the
modeling and design of computer-based systems in a way that demonstrates comprehension of
the tradeoffs involved in design choice (SLO j)
- Apply design and development principles in the construction of software systems of varying
complexity (SLO k)
Course Alignment Matrix
Directions: Assess how well Data Mining course contributes to the program’s student learning outcomes
by rating each course objective for that course with an I, P or D.
I=introduced (basic level of proficiency is expected)
P=practiced (proficient/intermediate level of proficiency is expected)
D=demonstrated (highest level/most advanced level of proficiency is expected)
NC – 9/29/05
SLO c
SLO d
SLO e
SLO f
SLO g
SLO h
SLO i
SLO j
SLO k
Understand the concepts, strategies, and
methodologies related to the design and
construction of data mining
Comprehend several data preprocessing
methods including the ChiMerge algorithm
Utilize data warehousing and OLAP
technologies
Apply appropriate mining techniques to
extract unexpected patterns and new rules
that are hidden in large databases
Understand current and future current data
mining technologies
SLO b
Course Objectives
SLO a
New Course Proposal – Page 7/8
P
P
I
I
P
I
I
I
I
P
P
I
I
P
I
I
I
I
I
I
D
D
I
I
P
I
I
I
I
I
D
P
P
I
I
P
I
I
I
I
I
D
D
P
I
I
I
I
I
I
I
I
P
I
I
15. Assessment of COURSE objectives
Course Objectives
Understand the concepts, strategies, and
methodologies related to the design and
construction of data mining
Comprehend several data preprocessing methods
including the ChiMerge algorithm
Utilize data warehousing and OLAP technologies
Apply appropriate mining techniques to extract
unexpected patterns and new rules that are hidden
in large databases
Understand current and future current data mining
technologies
Assessment
Exams and homework assignments
Exams and homework assignments
Exams, term papers, and group projects
Exams, term papers, and group projects
Exams, term papers, and group projects
17. Methods of Assessment for Measurable Student Learning Outcomes
A) Assessment Tools
The student learning outcomes will be assessed through a mix of homework assignments,
homework assignment presentations, a mid-term exam, a final exam, a term paper, a group project
and group project presentations.
B) Describe the procedure dept/program will use to ensure the faculty teaching the course will be
involved in the assessment process.
The course will put into place an assessment process that will involve all faculty teaching in the
program in the annual assessment of the program student learning outcomes. Each year few of the
student learning outcomes will be selected for assessment and the faculty teaching the course will
develop and implement an appropriate assessment plan.
NC – 9/29/05