This paper aims to identify potential influencers in a social network and find out whether they have impact on the sentiments, topics and stances of other user’s reviews.
The follower, following and review information of 5,106 users who post reviews for restaurants in Kusadasi, Turkey, was retrieved from Foursquare. Initially, the authors generated a social network by converting the following and follower data to a single large directed graph. Then, the potential influencers are uncovered by five different centrality metrics and the effect of these influencers on their followers written reviews are measured by three different the text mining approaches; sentiment analysis, topic modeling and stance detection.
This study reveals that influencers who can disseminate information more quickly act as bridges between other social actors, control the flow of information and have the highest quality and quantity of connections with other social actors, have a greater influence on their followers’ online review content compared with other influencers.
The limitation of this study is as in many online platforms, Foursquare reviews can be manipulated through fake reviews and this may result in the reliability of the findings. However, this limitation can appear in any online review platforms and is difficult to prevent.
The restaurant owners and managers can strategically identify and collaborate with the most suitable influencers who will increase their brand values, brand visibility and customer engagement. In addition, by understanding how influencers shape customer opinion and perceptions, they can manage their customer interactions by choosing the right influencers.
The implications of this study are both practical and theoretical. On the practical side, restaurant managers and owners can use insights gained from this study to select which influencers they need to collaborate with to enhance restaurant’s brand awareness. Theoretically, this study brings out and proves the effect of influencers on written online reviews of their followers.
In the theoretical side, unlike previous researches in the literature, this study focusing on information diffusion theory and finds out the effect of influencers on their followers’ written online restaurant reviews. This study performs both graph mining and text mining approaches to prove the hypothesis.
