AI2的限制

1. 程式積木一長就不好找
2. 無法redo／undo

Progress on questionaire design

• 4 research questions:

1. Paper-based learning sheet vs. none.
2. Electronic learning sheet vs. none.
3. Collaboration vs. none.
4. Trained vs. un-trained human summarizers.

4 samples learning objects also are picked up:

1. 新網釣技倆 Pharming => manually summarized.
2. 從一無所有走出來的 本土生菜女王
3. 免費交換 時間貨幣正流行
4. 電郵奇遇記 小女孩向俄總統要天竺鼠配對

Todo:

1. Continue to summarize 2-4.
2. Design assessment.
3. Try Google Questionaire.
4. What journal is good to submit?
5. Mail to Professor Chu.

• Research cooperation: Summarization for Chinese E-learning.

Research questions:

1. Psychological model.
2. Summarization model and instructional design.
3. Integration with Mind Mapping.

IA2S rebuilding

• Add execution-time evaluation with
  1. web version
  2. stand-alone version => try first


• Annotation tool survey:
  1. 符合SCORM2004標準且具學習策略數位教材之編輯工具的研發.ppt (done)
  2. 應用代理人技術於符合SCORM 標準且具學習策略之學習內容編輯工具的研究與開發.pdf (reference model)
  3. Visualized Online Simple Sequencing Authoring Tool for SCORM-compliant Content Package.pdf (reference model)
• Ask Ting-Hao's help for the IA2S (web version) execution-time evaluation
• Microsoft Agent tool

References

1. The Theory and Practice of Discourse Parsing and Summarization, 2000 (NTHU)
2. Advances in Automatic Text Summarization, 1999 (NTNU, NTU, NCTU)

Paper formulation

• Found a thesis related with the course structure. So, do a literachure review.



• Paper 1: Automatic Learning Object Annotation for Taiwan LOM Application Profile

- References: School For All

TKU, Reload Editor

Keyword generator

MS Summarizer

- Outlines: done.

• Paper 2: Computer Summary for On-line Chinese Learning

Paper reading: Operationalization of the Metadata Element “Difficulty”

• Organize OCAC's vocabulary set.
• Various formats and structures of a LO, and ambiguity and non-uniformed vocabulary in metadata elements

How to detect and construct the course structure?

• Read Tseng's paper again.
• Collect samples: w/ and w/o course structure
• Verify that keywords are the obvios clues or not?

Course analysis for OCAC

• Primary:

1. Chinese for Preschoolers--Children's
   Songs, Chinese's Saying, Chinese Riddles, and Chinese
   Phrases
2. Chinese Kindergarten
3. Children's Chinese Reader => No contents.
4. Learning Activites For Kids =>
   Points.
5. Chinese glossary E-Card

=> 1, 2, 5: Short contents. No summary is needed.

• Intermediate

1. Speaking Mandarin in 500 Words => No contents.
2. Speaking Mandarin in 1000 Words => Conversation.
3. Fun with Chinese Learning => Game-based.

• High-level

1. Chinese For Junior High School Level => Possible.
2. Chinese For High School Level => Possible.

------------------------------------------------------------------

• D/L "Possible" samples after go home.
• Get Stanford parser from CC.