Title: User response to e-WOM in social networks: how to predict a content influence in Twitter

Authors: Zohreh Yousefi Dahka; Nastaran Hajiheydari; Saeed Rouhani

Addresses: Faculty of Management, University of Tehran, Gisha Bridge, Chamran Highway, Tehran, Iran ' Sheffield University Management School, University of Sheffield, UK ' Faculty of Management, University of Tehran, Gisha Bridge, Chamran Highway, Tehran, Iran

Abstract: The purpose of this research is to find influential factors on electronic word-of-mouth effectiveness for e-retailers in Twitter social media, applying data mining and text mining techniques and through R programming language. The relationship between using hashtag, mention, media and link in the tweet content, length of the content, the time of being posted and the number of followers and followings with the influence of e-WOM is analysed. 48,129 tweets about two of the most famous American e-retailers, Amazon and eBay, are used as samples; results show a strong relationship between the number of followers, followings, the length of the content and the effectiveness of e-WOM and weaker relevance between having media and mention with e-WOM effectiveness on Twitter. Findings of this paper would help e-retailing marketers and managers to know their influential customers in social media channels for viral marketing purpose and advertising campaigns.

Keywords: electronic word-of-mouth; e-WOM; social media; e-retailing; content influence; data mining; Twitter; text mining.

DOI: 10.1504/IJIMA.2019.10024249

International Journal of Internet Marketing and Advertising, 2020 Vol.14 No.1, pp.91 - 111

Received: 10 Oct 2018
Accepted: 11 Nov 2018

Published online: 26 Mar 2020 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article