The Study on the Implementation of Course Scheduling System with Data Mining Techniques

• Main Authors: Shu-Ming Jen, 任書鳴
• Other Authors: Yao-Te Wang
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/54828891825209335966

碩士 === 靜宜大學 === 資訊管理學系研究所 === 95 === There are many variables are concerned in timetabling in the universities, colleges and institutes, including the teaching resources ,the classroom capacity, the curriculum characteristics, the teachers’ specialty, the hour limit, the willingness of holding a part-time post of the administrative duty and the preferred time interval of the class timetable and so on. All these variables can affect the result of course scheduling, and the result will then affect the daily activities of more than thousands of students and teachers in the next semester. In the past, timetabling solutions are always standing on the school’s position, and based on the planning of classes of departments and institutes. Key considerations are: faculty, equipments, society development, academic research and overall facilities of school. While the students are the most important users of course timetable, including their course registrations, learning performance and career planning, etc, the students’ demands are rarely seriously considered. When assigning courses to periods of class timetable, the purpose of course scheduler is to meet all the expectations from parties concerned: the arrangement of classes can conform to teachers’ requirements; all the students can take the courses they are interested in without the need to skip any class which is allocated in the same time interval as the ones they have already registered previously. This thesis focuses on the critical factors of timetabling from students’ points of view. Consider students’ behavior of course registration and their performance records in the first phase of course scheduling. The data mining technique is applied to history records of registration information to find out the associations among courses. Courses exhibiting strong association are allocated to different time slots in the class timetable to satisfy the students. In the mean time, by analyzing students’ past performances, the time intervals that students have better performance are reserved in class timetable with higher priority. The contributions of this thesis are: school can retain the classes for students which are unavailable previously due to time conflicts between classes; school resources can be used more efficiently since less staff and equipments investment are required. It’s possible to create a three-way win situation for school, teachers and students.