社會性影響與同類交往分析—以電信網絡為例

Analysis of Social Influence and Homophily Effect in Telecommunications Networks

宋長霖(Chang-Lin Sung)

指導教授 ： 謝志昇

國立臺灣大學/社會科學院/經濟學研究所/碩士(2021年)

https://doi.org/10.6342/NTU202102007

摘要

本論文量化電信網絡中手機使用行為的社會影響與同類交往。社會影響是指某人之行為受其同儕影響，進而導致兩人之行為越來越相似，概念等同「近朱者赤、近墨者黑」；而同類交往則代表某兩人因具有相似行為，造成此兩人建立關係、認識彼此，概念同「物以類聚」。本論文的資料來自中國某電信公司，包含中國西部某城市約 2.8 萬人之客戶通聯及手機使用紀錄。首先，我們使用 walktrap 演算法進行網絡分群，藉此得到數個高品質的網絡樣本。接著，應用隨機導向參與者模型估計每個網絡樣本的社會影響與同類交往。最後，利用「元分析」整合所有網絡樣本的估計結果，並揭示網絡結構如何促進社會影響。

關鍵字

社會影響 ； 同類交往 ； 隨機導向參與者模型 ； 元分析 ； Walktrap演算法 ； 網絡結構 ； 電信

並列摘要

This thesis tries to quantify the social influence on mobile phone usage and the homophily effect in telecommunications networks. Social influence refers to the concept that one’s behavior is influenced by peers’ behavior. While the homophily effect means that people with a similar behavior are more likely to be friends or form social ties. The data used in this thesis comes from a major mobile carrier
in a western province of China, which includes eight-month call detail records and monthly information of phone usage for each customer. We first apply the walktrap algorithm to cluster densely connected local networks. Then we use the stochastic actor-based model to estimate the social influence and homophily effect from network dynamics. Finally, we apply a meta analysis to summarize the estimation results from each network cluster and provide comprehensive insights on how network structural factors facilitate social influences.

並列關鍵字

social influence ； Homophily effect ； stochastic actor based model ； meta analysis ； walktrap algorithm ； network structure ； telecommunication

參考文獻

Abdi, Herve and Lynne J Williams (2010) “Principal component analysis,” ´ Wiley Interdisciplinary Reviews: Computational Statistics, 2 (4), 433–459. 39

Aral, Sinan and Christos Nicolaides (2017) “Exercise contagion in a global social network,” Nature Communications, 8 (1), 1–8. 37

Barrat, Alain, Marc Barthelemy, Romualdo Pastor-Satorras, and Alessandro Vespignani (2004) “The architecture of complex weighted networks,” Proceedings of the National Academy of Sciences, 101 (11), 3747–3752. 35

Becker, Joshua, Devon Brackbill, and Damon Centola (2017) “Network dynamics of social influence in the wisdom of crowds,” Proceedings of the National Academy of Sciences, 114 (26), E5070–E5076. 7

Bliss, Catherine A, Isabel M Kloumann, Kameron Decker Harris, Christopher M Danforth, and Peter Sheridan Dodds (2012) “Twitter reciprocal reply networks exhibit assortativity with respect to happiness,” Journal of Computational Science, 3 (5), 388–397. 36