282 MSCS -- Data Mining
Aug. 29, 2001
1
Syllabus MSCS 282: Data Mining
Fall 2001 Call #39254 http://www.mscs.mu.edu/~cstruble/class/mscs282/fall2001 Instructor: Craig A. Struble
Office: Office Hours: 386 Cudahy Hall 24 p.m. MW, and by appointment.
Class Meets: 5:458:25 p.m., Wednesday, CU118 Overview
"Data mining" refers to a collection of techniques for extracting "interesting" relationships and knowledge hidden in data. We will study a variety of these techniques and carry out practical exercises to understand what is and what is not "interesting." In the process, we will identify strengths and weaknesses of each technique.
Prerequisites
There are no formal prerequisites, but all students are required to have the instructor's consent. Background in at least one of databases, artificial intelligence, machine learning, pattern recognition, algorithms, or statistics will be helpful.
Textbook and References
Required
Jiawei Han and Micheline Kamber, Data Mining: Concepts and Techniques, 2001, Morgan Kaufmann Publishers, ISBN 1-55860-489-8. Communications of the ACM, Special Issue on Data Mining, November 1996. Available online through the library. Communications of the ACM, Special Issue on Knowledge Discovery, November 1999. Available online through the library. Additional material will be on reserve in the library and on the course web site.
MSCS 282 -- Data Mining
Aug. 29, 2001
2
Grading
Your grade will consist of the following components, weighted as shown: Homework Term Project 50% 50%
Final letter grades will depend on the overall performance of the class. The table below shows the overall percentage you must acheive to guarantee a given letter grade. Percentage 92% 88% 82% 78% 72% 68% 60% Letter Grade A AB B BC C CD D
Writing Expectations
Good writing skills are essential for effective communication of your ideas. All work submitted in this class is expected to be in well written English. Poorly written work may have points deducted, worth no more than 5% of the assignment grade. I will correct or point out grammatical and spelling errors in your work. In extreme cases, the work will be returned to be rewritten and resubmitted, with a 10% penalty. If you are not confident in your ability to write English well, you should seek help from the Ott Memorial Writing Center. Visit http://www.marquette.edu/writingcenter/ for more information.
Homework Assignments
Homework assignments will primarily consist of discussion questions and experiments related to data mining. You will be given 12 weeks to complete a homework assignment. Homework assignments must be submitted in class the day the assignment is due and prepared electronically (e.g., using word processing software). Each assignment is graded on a 100 point scale.
Term Project
The term project is a group project intended to tie together and concepts techniques discussed in class. Groups may consist of up to 3 students. Your group will write a project proposal, maintain a project web site, carry out the proposed project, and submit a final project report of no more than 15 pages in length. The goal of the project is to produce a report that could be published in a regional conference about data mining or some other trade publication (e.g., Dr. Dobbs). More details about the project will be made available in lecture and on the course web site. The major milestones for the project are listed below.
MSCS 282 -- Data Mining Date September 12, 2001 October 3, 2001 October 17, 2001 November 14, 2001 December 5, 2001 December 7, 2001
Aug. 29, 2001 Event Groups Selected Draft Proposal Due Final Proposal Due Progress Report Due Class Presentation Project Due
3
Attendance Policy
Each student is expected to attend every lecture. If you miss two lectures, you will be penalized a half letter grade and formally warned of the possibility of receiving a WA (Withdrawal for Excessive Absences) If you miss three lectures, your final grade will be penalized a full letter grade. Missing the fourth lecture will result in being assigned a WA, which cannot be changed to a W. In addition, extreme tardiness to class may be recorded as an absence.
Academic Honesty
All students are expected to adhere to the standards of student conduct as described in the Community Expectations section of the student handbook. Homework assignments are intended to reflect individual effort in the course. Students may discuss homework assignments in a general way; i.e., discussing the nature of the assignment or providing clarifications. Sharing homework solutions in any f...