2024-03-28T12:38:13Zhttps://gredos.usal.es/oai/requestoai:gredos.usal.es:10366/1342682022-02-07T15:34:17Zcom_10366_122575com_10366_4512com_10366_3823col_10366_134243
Lahuerta Otero, Eva
Cordero Gutiérrez, Rebeca
2017-09-05T10:59:08Z
2017-09-05T10:59:08Z
2016
Computers in human behavior. (64), pp. 575–583.
http://hdl.handle.net/10366/134268
http://www.sciencedirect.com/science/article/pii/S0747563216305258
The purpose of this study is to investigate influencers on Twitter to discover the characteristics of their tweets through PIAR, a unique data mining research tool developed by the University of Salamanca that combines graph theory and social influence theory. An analysis of 3853 users posting about two automotive Japanese car firms, Toyota and Nissan, reveals the characteristics influencers have on this social network. The findings suggest that influencers use more hashtags and mentions on average when they tweet, and their word count is fewer than those with less power on this virtual community. Surprisingly, they tend to include less embedded links on their posts. Additionally, influencers have on average a large number of people they follow and they clearly express their opinions and feelings (either positive or negative) when tweeting. The results broaden the understanding of how influencers write and behave on social networks when they communicate with their users' community. Further, it provides insights for practitioners and marketers on how to discover influencers talking about their brands by observing tweets' content.
en
info:eu-repo/semantics/openAccess
Computer Science
Looking for the perfect tweet. The use of data mining techniques to find influencers on twitter
info:eu-repo/semantics/article