Validation of the effects of influencers in social networks through disseminating the facebook application

Abstract

This research proposes and evaluates a methodology for discovering influencers within online social networking sites (OSNSs), which provide powerful means of sharing common interests and communicating between users. Finding outstanding influencers out is major perspective in success of viral marketing. By using data from Facebook, one of the most popular online social networking sites, we validate the effects of influencers extracted by the methodology. Proposed methodology for discovering influencers consists of two-step process. In step one, structural analysis based on the social graph which connects users who are friends each other is performed. Through the degree centrality computation, we choose the influencer candidate subgroup. Next, the second step is activity analysis using user’s activity history. After whole process, we finally obtain a few powerful influencers who can affect more effectively other people than mediocrities. The positive effects of influencers determined from the methodology are confirmed by experiments that disseminate flash games on Facebook. Our research lays the groundwork for searching influencers within online social networking sites. And proposed methodology contributes to competitive strategies in viral marketing.