應用社會網路與社群情感分析於網路口碑對銷售的影響

Network Effects and Social Sentiment: The Influence of Electronic Word-of-Mouth on Sales

白昆庭(Kun-Ting Pai)

指導教授 ： 王俊程

國立清華大學/科技管理學院/服務科學研究所/碩士(2013年)

摘要

應用社會網路與社群情感分析
於網路口碑對銷售的影響

研究生：白昆庭 指導教授：王俊程 博士

論文摘要
網路中社群媒體使用者所形成的網路口碑對於消費者購買決策的影響越來越被重視。藉由計算社群媒體中使用者創造的內容可以得到社群中消費者的情報，而且有越來越多的企業將此方法視為新型態的行銷工具。本研究以網路上手機討論網站作為資料來源，結合產品實際的銷售資料作分析，包含主題數量、回應數量、社會網絡結構指標以及內文情感的散佈等構面來探討網路口碑的形成以及對於產品實際銷售的影響力。
本研究發現網路口碑中的主題對於產品銷售的影響是正向的，回應則是對於口碑形成的有正向影響。社會網路分析的指標中，網絡密度和程度中心性對於產品銷售是負向影響，但是中介中心性和群聚係數對於銷售是正向影響；然而在分析口碑形成中，網路密度和程度中心性的影響不一致，密度對於口碑形成是負向影響，程度中心性則是正向影響。在網路口碑內文中的情緒散佈，結果顯示情緒散佈越混亂對於產品銷售和口碑形成都有正向的影響。以上分析結果可當成企業在制定網路口碑的行銷策略時重要的參考指標。

關鍵字：網路口碑 (Electronic Word-of-Mouth)、 社會網路分析 (Social Network Analysis)、 社群情感分析 (Social Sentiment Analysis)

關鍵字

網路口碑 ； 社會網路分析 ； 社群情感分析

並列摘要

Network Effects and Social Sentiment:
The Influence of Electronic Word-of-Mouth on Sales

Author: Kun-Ting Pai Advisor: Dr. Jyun-Cheng Wang

Abstract
Electronic word-of-mouth (eWOM) from online social media has become increasingly important in influencing customers’ purchasing decisions. More and more firms are taking advantage of eWOM as a new marketing tool, which can collect social intelligence data over different platform content. This study uses the data from the online mobile community as well as actual sales data to analyze the impact of topic, response, social network indicators, and eWOM entropy to determine eWOM generation in relation to product sales.
The results show that the topic of eWOM has a positive impact on product sales, and the topic responses of eWOM have a positive impact on eWOM generation. In social network analysis, degree density and degree centralization have a negative impact on product sales, but betweenness centralization and the clustering coefficient have a positive impact on product sales. Degree density and degree centralization have inconsistent results in analyzing eWOM generation. Degree density has a negative impact on eWOM generation, and degree centralization has a positive impact on eWOM generation. This result also shows that the dispersion of social sentiment has both a positive impact on product sales and on eWOM generation. In practice, these results could be referenced for companies to establish future marketing strategies in social commerce.
Keywords: Electronic Word-of-Mouth, Social Network Analysis, Social Sentiment Analysis

並列關鍵字

Electronic Word-of-Mouth ； Social Network analysis ； Social Sentiment Analysis

參考文獻

Liu, Yang, Huang, Xiangji, An, Aijun, & Yu, Xiaohui. (2007). ARSA: a sentiment-aware model for predicting sales performance using blogs. Paper presented at the Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval.

Tang, Chih-Ching. (2005). A study of the correlation among online community and market status–using cellular phone as example. National Chung Cheng University.

Huang, C. –L., Chung, C. K., Hui, N., Lin, Y. –C., Seih, Y. –T., Chen, W. -C., Lam, B., Bond. M., & Pennebaker, J. W.. (2012). The development of the Chinese Linguistic Inquiry and Word Count dictionary . Chinese Journal of Psychology 54(2), 185-201.(in Chinese)

Berger, Jonah, Sorensen, Alan T, & Rasmussen, Scott J. (2010). Positive effects of negative publicity: when negative reviews increase sales. Marketing Science, 29(5), 815-827.

Bounie, David, Bourreau, Marc, Gensollen, Michel, & Waelbroeck, Patrick. (2008). Do Online Customer Reviews Matter? Evidence from the Video Game Industry.