符合SCORM標準之學習管理系統的適性化學習物件編組

Adaptive Orchestration of Learning Objects in the SCORM-Compliant Learning Management Systems

廖界信(Chieh-Hsin Liao)

指導教授 ： 廖慶榮

中原大學/商學院/資訊管理研究所/碩士(2007年)

https://doi.org/10.6840/cycu200700309

摘要

摘要
數位學習為了能夠滿足學習者對於課程先備知識差異上的學習需求，因此學習管理系統推薦給學習者的教材與次序安排是一個重要的影響因素。

本研究使用資料探勘的學習者分群與數量化關聯法則演算法，來分析學習優異的學習者之學習方法，藉此找出適合學習者的推薦學習路徑。學習者透過本系統的推薦路徑可以較容易達成學習目標。

本研究所需之學習管理系統修改自SCORM 2004 RTE，以期符合所需要的相關資料紀錄。課程製作與包裝使用Reload Editor 2004與Camtaisa Studio來進行。使用最大期望值EM分群演算法進行學習者分群，以LqiTid數量化關連法則探勘演算法進行學習路徑的探勘。為了驗證推薦路徑的可行性，透過實驗設計安排以準實驗設計之「不等的前測–後測設計」方式，設計實驗組與對照組的實驗對照來探討學習紀錄與學期成績在本研究的課程學習上，提供探勘與推薦機制對學習者學習的成效。實驗結果顯示，結合數位學習與本研究所提出的推薦機制，使得學習者在學習上確實能夠得到適性化的課程推薦幫助，學習成效優於沒有推薦機制的對照組。

本研究提出的方法與系統應用在個人化學習服務（Personalized Learning Services）的確能夠提供適性化的課程推薦，來滿足不同學習者之間的差異。藉此以提供學習者最適合的學習路徑與課程，達到適性化的學習服務組合的目的。

關鍵字

SCORM ； 資料探勘 ； 適性化

並列摘要

ABSTRACT

In order to satisfy the learning needs due to the differences between every learner’s prior knowledge, the course contents and proper arrangements in sequence which are recommended by the learning management system definitely plays a critical role in e-learning.

The research proposes utilizes data mining’s learner-clustering and quantitative relation algorithm methodologies. The system mines a recommendation learning path which will be relevant for the learner and it is also anticipated that the learner can accomplish the learning objective by following the recommendation learning path.

The Learning Management System in this research was modified from SCORM 2004 RTE to record the relevant data. Course arrangement was using Reload Editor 2004 and Camtaisa Studio to complete. The research also utilizes EM algorithm to perform the learner-clustering and LqiTid algorithm to mine the learning path. In order to verify the feasibility of recommend path, the research used quasi-experimental design, a practical experiment which is performed by Experimental and Comparison Group running on the learning system. Furthermore, this research will analyze each learner’s learning records and results to discuss whether the system helps learners on the mining and recommendation of adaptive learning path or not.
The experimental results show that combine the e-learning and proposed model helps learners to learn and can provide adaptive recommendation course contents to disappear the gaps between different learners, and the relevant is better than Comparison Group.

The research is also anticipated that the proposed system which is utilizes on Personalized Learning Services can provide adaptive recommendation course contents to disappear the gaps between different learners.

並列關鍵字

Data Mining ； Adaptive ； SCORM

參考文獻

S.-M. Tasi, and C.-M. Chen, “Mining Quantitative Association Rules in a Large Database of Sales Transactions,” Proc. 17th Journal of Information Sciences and Engineering(JISE ’01), 2001, pp 667-681.

Advanced Distributed Learning (ADL), “Sharable Content Object Reference Model (SCORM) 2004 3nd Edition Overview,” 16 Nov. 2006; http://www.adlnet.org/.

O. Bohl, J. Schellhase, and U. Winand, “The Sharable Content Object Reference Model (SCORM) - a critical review,” Proc. Int’l Conf. on Computers in Education (ICCE’02), 2002, pp. 950-951.

J. Broisin and P. Vidal, “A Management Framework to Recommend and Review Learning Objects in a Web-based Learning Environment,” Proc. 6th IEEE Int’l Conf. on Advanced Learning Technologies (ICALT’06), 2006, pp. 16-18.

U. Fayyad, G. Piatetsky-Shapiro and P. Smyth, “From Data Mining to Knowledge Discovery in Databases,” American Association for Artificial Intelligence, 1996, pp. 37-54.