遠距教學提供教師及學習者一個更便利的學習環境,舉凡傳統課堂學習能夠做到的事,都可以以現代化的網路學習系統來完成,視訊及音訊搭配超媒體的使用,將呆版的傳統課程變成豐富有趣的線上學習課程。儘管如此,網路課程仍有其無法取代傳統課程之處,因為網路課程的學習不受時間空間的限制,所以無法確實掌握學生的學習情形,也無法藉由線上測驗來評量學生的學習成果。 本論文所提出的線上學習活動診斷系統,輔助教師去定義線上學習系統資料庫中學生的學習歷程(Learning portfolio),並設計了簡易的使用介面,幫助教師對定義好的學習歷程進行分析,同時建立學生的學習行為模組。教師可以將舊課程的學習行為模組套用在進行中的課程上,預測修課學生未來可能會產生的學習情況,並給予指導及輔助。系統內並包含自動監督機制,可依照教師設定定期偵測學生的線上學習情況,並對處於特定學習情況的學生發送警訊。在完整掌握學生的學習情況後,我們也設計了以學習歷程為依據的評量機制,輔助教師依照學生的學習歷程給予適當的評量值。 本論文提出了一個高自由度的研究平台,分析不同時期的資料記錄能夠觀察學生在課程中所產生的行為變化,而將學習歷程以結構化學習特徵(Structuralized Learning Feature,SLF)定義的手法,更使得本系統具有跨學習平台的分析優勢。
Distance learning provides a more convenient way of learning to teacher and learner. All things that can be accomplished by traditional learning can also be done by e-learning system. By using multimedia technologies such as video, it is possible to turn traditional course still have restriction to replace traditional courses. Unfortunately, e-learning course still have restriction to replace traditional courses. Because e-learning is not restricted to time and space, it is hard to understand students’ learning situation and also hard to evaluate learning achievement correctly by on-line examination. In this thesis, a system which applied characteristics described is proposed to diagnosis of e-learning, in this system, we help teachers to describe the learning help teachers to define and model the student behavior modules. Teachers could adapt the past courses behavior modules to current courses. By using this, Teachers can try to predict the students’ learning situation and offer assistance appropriately for those students who may failed the courses. The system also have an auto-supervising mechanism that can monitor student’ learning situation on a regular time schedule set by teachers and send earning messages for those students who have unusual learning behavior. When student’ learning situation are known well, the system can help teachers to evaluate students by their learning profolio. To sum up, this thesis has proposed a system with high degree of freedom for research. We can analyze the profolio from different periods to observe the changes of learning behavior. By using Structuralized Learning Feature(SLF),it is possible to advantage the system crossing different learning platforms.