基於互補度與社群網路分析於基因演算法之分組機制;Grouping based on Complementary Degree and Social Network Analysis in Genetic Algorithm

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Computer Science and Information Engineering > Electronic Thesis & Dissertation > Item 987654321/65590

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/65590

Title:	基於互補度與社群網路分析於基因演算法之分組機制;Grouping based on Complementary Degree and Social Network Analysis in Genetic Algorithm
Authors:	蘇皇名;Su,Huang-Ming
Contributors:	資訊工程學系
Keywords:	合作式學習;分組;基因演算法;社群關係;Cooperative Learning;Grouping;Genetic Algorithm;Social Network
Date:	2014-07-10
Issue Date:	2014-10-15 17:05:37 (UTC+8)
Publisher:	國立中央大學
Abstract:	在過去幾年中，合作式學習已成為最重要的教學策略之一。在合作學習裡，如何將學習者們適當地分組成為了一個重要的議題。為了解決這個問題，許多學者提出了各式各樣的方法。在本研究中，我們採用一種不同的方法，利用學習者們的知識結構互補程度以及社群網路結構來提高學習者之間的互動以及團隊合作。另外，本研究利用基因演算法(GA)來演化出更好的分組結果。另一方面，本研究開發出一套系統，學習者以及教師可以利用數位學習系統來快速查詢分組結果以及取得課堂上的訊息。而分組系統可以記錄每一位學習者的學習狀態以及與組員之間的互動程度。分組系統利用這些資訊來動態調整每一次的分組結果。在研究結果中，我們可以發現本研究的方法確實可以優化分組。本研究可以在維持高異質分組的狀況下，同時兼顧分組結果的接受程度。另外，學生利用系統分組進行合作學習在學習效果上有明顯提升。經過實驗結果分析後，我們發現實驗組的後測平均分數高於對照組，而且實驗組的知識結構比對照組要來的一致。在最後的問卷中，學生說他們喜歡系統上的分組功能以及分組結果。本研究所採用的分組方法確實能夠幫助他們學習。 ;In the past years, Cooperative Learning has become one of the most important teaching strategies. Helping learners grouping appropriately is now becoming more and more important. To solve the problem, a lot of methods have been proposed. In this study, we employ a novel grouping approach that considers the complementary degree of learner’s learning state and social networks to enhance interaction and teamwork between learners. This study also used genetic algorithm (GA) to generate better grouping results. Moreover, a set of systems have been developed. The e-learning system was developed for learners and tutors that they can view the grouping result and academic information conveniently. In grouping system, it will record learners’ learning statuses and interaction between team members to adjust grouping result from each assignment dynamically. In the end, the results show that the proposed approach can optimize the grouping well. The proposed method of grouping can generate high heterogeneous grouping results and the learners are satisfied with the grouping results at the same time. Meanwhile, the learners’ learning effects are improved by using the proposed method of grouping in cooperative learning. The mean score of post-test in the experimental group was higher than the control group. Moreover, the experimental group learners’ academic level reaching more consistent than the learners of the control groups. Finally, the learners said that they liked the grouping function of the system in their feedback. The grouping method of the system really helped them to learn efficiently.
Appears in Collections:	[Graduate Institute of Computer Science and Information Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	448	View/Open

社群 sharing

Loading...