設計一問題本位學習小組討論之智慧型代理人─以資料庫管理課程為例

Designing an intelligent agent for Problem─based learning group discussion ─ Using Database Management System Course as Example

胡哲豪(Zhe-Hao Hu)

中原大學/工學院/資訊工程研究所/碩士(2011年)

https://doi.org/10.6840/cycu201100914

摘要

在問題本位學習(Problem-based leaning)中，小組討論是一個重要的歷程。學生可以藉由小組討論的方式，來獲得知識，並且學習解決問題的方法。在小組討論中，教師不僅是知識的傳遞者，更是討論的誘導者。
但是，在一個有許多個小組的環境之下，勢必要有多個教師來協助學生進行討論。在此情況下，勢必會造成教師的負擔。本研究設計了一智慧型代理人來協助教師。為了要協助教師引導學生討論，此智慧型代理人必須要有分析，並且瞭解學生討論內容的能力。因此，本研究嘗試去建立討論室中的概念圖(Concept Map)，並且依據此概念圖來評斷學生對學科的瞭解程度，進而給予適當地建議。
此代理人被應用在一大學的資料庫管理(DataBase Management)課程。在實驗中，學生們在Moodle-課程教學網站上利用討論室討論。同時，此代理人也參與了每一組的討論，並且給予討論建議。在課程結束後，學生們被要求填寫科技接受度問卷(technology acceptance model)，以便調查此種方法是否能夠幫助學生討論。最後，依據收集到的資料進行分析，並加以討論，來得知藉由代理人在討論中給予的方式是否能提昇討論成果。

關鍵字

概念圖 ； 問體本位學習 ； 概念圖評分 ； 概念圖自動產生

並列摘要

In problem-based learning (PBL), small group discussion is the crucial factor for the effectiveness of the learning process. It is necessary for the discussion to be monitored and guided by the instructor for most e-PBL implementations. Overmuch groups cause lack of instructors is evidently. This research purposed an intelligent agent-IPBLA (intelligent problem-based learning assistant) with ability of discussion contents analyzing/evaluating, students activity evaluating and discussion promoting. We expect IPBLA could reduce effort of instructor who conducts excess groups and further students’ discussion. In implementation, IPBLA participates in group discussion and provides suggestions to students.
In this research, text matching is adapted to build concept map of students automatically. For evaluating students’ knowledge level, this concept map is scored. Furthermore, IPBLA gives suggestions to students according to the result of scoring.
For validating, IPBLA is adapted to a college-level database management system course under e-PBL environment. In experiment, students are asked to discuss two independent questions on Moodle - online learning environment of this course sequentially. Certainly, IPBLA joins group discussions of students. After group discussion, students are requested to accomplish self assessment and peerassessment. In addition, students complete TAM (technology acceptance model) questionnaire after last group discussion. Although the increasing of students’ knowledge level is not established, most students agree that IPBLA is useful for discussing.

並列關鍵字

Concept Map ； Concept Map scoring ； Concept-Map mining ； Problem-based learning

參考文獻

游文楓（2003）。網路化問題解決教學策略對學生生物學習成效的影響。未出版之碩士論文，國立交通大學理學院網路學習，新竹。

陳冠廷(2005)。“Designing a porblem-based Learning Conversation Agent based on Speech Act Theory“。中原大學資訊工程研究所碩士論文。

S.-M. Bai & S.-M. Chen, “Automatically constructing concept maps based on fuzzy rules for adapting learning systems”, Expert Systems with Applications, Vol. 35, No(1-2), July-August 2008, pp. 41-49

B. E. Cline, C. C. Brewster & R. D. Fell , “A rule-based system for automatically evaluating student concept maps”, Expert System with Applications, Vol. 37, No 3, 15 March 2010, pp. 2282-2291

N.-S. Chen, Kinshuk, C.-W. Wei & H-.J Chen, (2008). “Mining e-Learning domain concept map from academic articles”. Computers & Education. Vol. 50, No 3, pp 1009-1021