運用資料探勘建立個人適性化學習網站

Using Data Mining Techniques to Build Adaptive E-Learning Web Site

黃星峯(Hsing-Feng Huang)

指導教授 ： 陳煇煌

大同大學/經營設計學院/資訊經營學系所/碩士(2004年)

摘要

於現今多數的線上教學網站中，其課程內容架構大致都在建立課程時，都早已經做好其相對應的課程教材連結，不論進入網站的學習者是誰，幾乎都提供一致性的連結式樣，或者是教材內容因時間拉長，不斷加入課程資料，導致學習者迷失在這錯綜複雜的教材連結中，因此學者Conklin 說過『迷失方向(Disorientation)和認知負荷(Cognitive Overhead)是超媒體文件中最主要的兩大問題』[1]。
因此本研究著重發展一適性模式，藉由實施教材前的前測測驗、瀏覽教材的行為以及閱讀教材結束之後測測驗歷程，將這些資料匯聚建立資料倉儲，並運用資料探勘技術-分類法則與關連法則，分析建立群體航行模式，並且針對瀏覽行為資料建立個人航行模式，緊接著利用個人航行模式與群體航行模式進行對映，找出個人航行樣式，最後利用個人航行式樣預測使用者可能瀏覽動作，並予以適性指引。

關鍵字

線上學習網站 ； 網際網路挖掘 ； 資料挖掘

並列摘要

The majority of e-learning Web sites have predefined course frameworks. No matter who enters the Web site, almost the same link types are offered. The course materials are added when the time is prolonging. As a result, the learners can be lost in the intricate links of teaching materials. Conklin indicated that ‘disorientation’ and ‘cognitive overhead’ are the two prime issues in hypermedia documents [1].
Thus, this thesis research focuses on the development of an adaptive model by taking pre-learning test before enrolling in the course materials, recording the browsing behavior on the course materials, and taking post-learning exam at the end of learning the course materials. These data are assembled to set up a data warehouse. We use data mining techniques, classification and association, to analyze the collected data to set up group navigation model and personal navigation model. Finally, we use the group navigation model and personal navigation model to predict learner’s personal navigation pattern and give adaptive guidance to the learner.

並列關鍵字

Data Mining ； E-Learning Web site ； Web Mining

參考文獻

[1] Conklin, J., “Hypertext: An introduction and survey”, IEEE Computer, September 1987, pp. 17-41.

[2] Perkowitz, M. and E. Oren, “Adaptive Web Sites: an AI Challenge”, In Proceedings of the 15th International Joint Discussion on Artificial Intelligence, Nagoya, Japan, 1997, pp. 16-23

[3] Perkowitz, M. and E. Oren, “Towards Adaptive Web Sites: Conceptual Framework and Case Study”, 1999, http://www8.org/w8-papers/2b-customizing/towards/towards.html

[4] Brusilovsky, P., “Methods and Techniques of Adaptive Hypermedia”, User Modeling and User-Adapted Interaction 6, 1996, pp. 87-129.

[5] Bra, D.B., P. Brusilovsky, and G.J. Houben, “Adaptive Hypermedia: From Systems to Framework”, ACM Computing Surveys’ Electronic Symposium on Hypertext and Hypermedia, Vol. 31 No.4es, December 1999, Article No. 12.