| Summary: | 碩士 === 臺南師範學院 === 資訊教育研究所 === 88 === As the rapid progress on the World Wide Web, it has become a natural trend by learning through the Internet. However, instructors may not be able to understand the learning statuses of students based an active web based learning environment. Therefore, it has become an critical issue to support an effective and efficient tool for managing students'' learning processes, such that the instructors can evaluate the students'' actual learning processes and promote the students'' learning achievements by dynamically adjusting the courses web structures. Basically, there are some data that can be collected via a typical Internet management tool, such as, what web courses the students browsed, including their corresponding browsing frequencies, browsing time, browsed URL, etc., which called learning sequential data. However, those data only present the learners’ processes, but there are no useful information which have been well analyzed as useful learning patterns. (These patterns may include web learning units and web browsing path.) The understanding of the students'' learning pattern enable the instructors to precisely know which web pages the students usually visited and how they learned. In this these, we will provide an efficient approach to analyzing the collected data to learning patterns. A new efficient data mining technology will be proposed and applied to discover the associations among those information components automatically. From experiment result, we find our approach has better performance than the others.
|
|---|
