以MDG (MLCs Directed Graph)為基礎來尋求一多媒體教材素材的品質管制方法

A Study of a Quality Control Method - Based on MDG (MLCs Directed Graph)

李仲智(Chung-Chih Lee)

指導教授 ： 陳登吉

國立交通大學/資訊學院/資訊學院碩士在職專班資訊組/碩士(2010年)

https://doi.org/10.6842/NCTU.2010.00870

摘要

多媒體教材日漸普及，其品質也益受重視，而目前大多以其內容、創意為品管的目標，但製作過程就比較少被討論。本論文希望能在製作過程中，以軟體工程中的Graph為基礎，利用其現有的理論及演算法，希望能幫助多媒體教材在製作過程的錯誤減少，進而使的品質能夠提昇。藉由定義MLCs Directed Graph (MDG)，來達成將教材及素材間的關連建立。它將多媒體教材的組成素材視為節點(nodes)，代表多個素材節點組成的邏輯群組也是一個節點；節點與節點間關連以邊(Directed edges)串連形成Directed Graph (Digraph)，而此定義此關連的規範，是來自於依其自然語言需求而產生的「需求表」，進而依其需求表，將教材與素材的關連轉換成Digraph。一份教材也視為一個節點，以它為根節點(root node)，依其邊(directed edges)的關連可以找出所有相關連的節點，這就是一個MDG。
有此MDG之後，便可以尋找多媒體教材製作過程可能發生的錯誤或品質不良事件，找出這些事件發生的模式，利用Graph中的理論及方法，可在系統內自動化檢驗出來，冀以將之排除，使其在製作過程的錯誤減少，以達品質提昇的目的。本研究歸納出以下九大點的品質不良事件：一、利用MDG的關連結構，可將教材內散亂無章的素材串連起來，並方便素材的管理。二、參考MDG的關連，比較容易對照相關素材的正確性。三、教材內或邏輯群組內的素材，可有系統地做整備檢查，防止素材的缺漏。四、當素材面臨修改變更時，由MDG可以找出與之相關的節點，以利週邊效應的評估。五、可利用MDG來標記出沒有被使用到的素材節點。六、設計階段的元件缺漏檢查。七、基本的語意檢查。八、MDG可定義群組關係與節點關連，藉此可做到語法外的元件群組關連檢查。九、可以發現元件間的遞迴。以MDG為基礎，冀望未來有更多可能的發展及研究方向；例如歸納出更多的品質不良事件、進階的語法語意檢查、無限廻圈的檢查、Valid & Validation及反向工程等等。

關鍵字

MDG ； Graph ； 多媒體 ； 教材 ； 品質 ； 管制

並列摘要

Multimedia Learning Contents (MLCs) become popular and quality of it is also becoming important. But most of them focus on content quality, few taking care on its production process quality issues. This study wishes to focus on process control issues of MLCs. We want to use Graph theories to help reducing errors happening during MLCs producing.
This study defines a new term ‘MLCs Directed Graph (MDG)’, and takes care on those materials of MLCs and builds relations on materials to MLCs or materials. The build-up materials of MLCs are seen as nodes, and a logical organization of material nodes could also be seen as a node. Nodes and nodes are linked by directed edges and which forms a directed graph (Digraph). It comes from demands of natural language, and translates it to a digraph. A MLC itself could be seen as a root which is a node, too. By searching those edges related to a root we could find a Digraph, and this is an MDG.
With this MDG, we could find and define MLCs quality defection issues, finding samples of those issues, and prevent form happening of these issues by using graph and graph mythologies. And this will improve the quality of MLCs by reducing defection issues. This study concludes 9 defection issues: (1) Materials in chaos and without management. MDG could build a structure of materials and easy to management. (2) With edge relations and MDG, it is easier to check the correction of materials. (3) Integrity of materials of MLCs or logical organizations could be checked easily and systematically. (4) When changes happen, side affections could be evaluated easily by reviewing nodes related. (5) Unused nodes could be noted by MDG. (6) Lacking material nodes check by MDG at design time. (7) Basic syntax error check. (8) Extra group relation defined between nodes helps node group relation examinations beyond syntax. (9) Looping could be marked.
iii
Basing on MDG, there are lots things could be dig in. Maybe more quality defection issues could be found. Further syntax error find out. Infinite loop find out at design. Valid & Validation and reversing engineer may work out.

並列關鍵字

MDG ； Graph ； multimedia ； learning ； content ； quality control

參考文獻

[1] W. D. Wallis, A Beginner's Guide to Graph Theory, 2nd edition, Birkhauser, Boston, June 8, 2007.

[4] Advanced Distribute Learning (ADL) Initiative, Sharable Content Object Reference Model (SCORM) 2004 4th Edition: Overview, Version 1.0, March 31, 2009.

[9] IEEE, IEEE 1484.11.2 Standard for Learning Technology – ECMAScript Application Programming Interface for Content to Runtime Services Communication. November 10, 2003, Available at: http://www.ieee.org/

[16] Qing Li, et al., Conceptual Modeling - ER 2008, Part II, Barcelona, Spain, November 17, 2008.

[17] Cormen, Thomas H. Leiserson, Charles Eric. Rivest, Ronald L., Introduction to algorithms, Cambridge, Mass.: MIT Press, 2001.