基於社會網路分析之學習者行為探勘

Learning Behaviour Mining Based on Social Network Analysis

覃霆輝(Teng-Fai Seum)

指導教授 ： 陳恆佑

共同指導教授 ： 張克寧(Keh-Ning Chang)

國立暨南國際大學/科技學院/資訊工程學系/碩士(2014年)

摘要

本研究主要以基於社會網絡分析（SNA）開發資料挖掘技術，觀察學生在Facebook平臺上進行合作學習時對社交學習行為有何影響。本研究共推導出七種類型的SNA指標，以探討學生個人和群體之間的溝通模式：參與度、領導性、互動性、網絡離中性、訊息掌握度、貢獻度以及迴響度。為了證明本研究的SNA指標的可行性，我們進行了一項實驗。在實驗中，參與者被分為三種基於性別組合的討論群：（1）混合性別的討論小組，（2）只有女性的討論小組，和（3）只有男性的討論小組。本研究根據不同的性別組合在組內或跨組的討論活動進行分析，以觀察參與者的性別配對和社交學習行為有和關係。
實驗結果顯示，女生在同性組合下產生的學習效果比在異性組合下更優異；男生在異性組合下產生的學習效果比在同性組合下更優異；同一性別的組合產生的學習效果比兩種性別混合的組合表現優異。

關鍵字

社會網絡分析 ； 學習者行為分析 ； 性別組合

並列摘要

This study developed mining algorithms based on social network analysis (SNA) to investigate the effects of gender and gender paring on students’ social learning behaviors on Facebook. Seven types of SNA indexes were derived to discuss both individual and group communications: engagement, leadership, interaction, dispersion, broker, contribution, and response. We also conducted an empirical study to prove the feasibility of the proposed algorithms. In the experiment, participants were grouped into three types of discussion groups based on gender combinations: (1) the mixed-gender discussion group, (2) the girl discussion group, and (3) the boy discussion group. Social learning behaviors, such as in-group communication and between-group communication for different gender combinations were then analyzed to see the gender and gender-paring effects.
Experimental result shows that the single-gender groups had better learning performance than the mixed-gender groups. Besides, females in the single-gender groups performed better than in the mixed-gender groups; males could perform better when they were in the mixed-gender groups.

並列關鍵字

Social network analysis ； Gender composition ； Learning behaviour

參考文獻

參考文獻

[ 1 ] Ahn, Y.-Y., Ahnert, S. E., Bagrow, J. P., & Barabasi, A.-L. (2011). Flavor network and the principles of food pairing. Scientific Reports, 1, 196. doi:10.1038/srep00196.

[ 2 ] Aviv, R., Erlich, Z., & Ravid, G. (2003). Network Analysis of Interactivity in Online Communities. Association of Internet Researchers Fourth Annual Conference. Toronto, Canada.

[ 3 ] Bernard, H., Killworth, P., & Sailer, L. (1982). Information accuracy in social network data V: An experimental attempt to predict actual communication from recall data. Social Science Research, 11, 30-66.

[ 4 ] Cho, H., Gay, G., Davidson, B., & Ingraffea, A. (2007). Social networks, communication styles, and learning performance in a CSCL community. Computers and Education, 49(2), 309-329.