How do fandom types differ? A taxonomy of K-pop fandom with network embedding

Since fandom as a distinctive customer group known for their exceptional loyalty and profitability has increasingly gained influence, fandom marketing has become crucial for the success of businesses. However, there is a lack of quantitative methods to identify and classify different types of fandoms. This study proposes a network embedding-based approach to identifying different types of fandoms. The proposed approach is applied to the K-pop industry, which has experienced remarkable growth due to its fandom. A fandom of a K-pop artist is captured as a network of Twitter users based on their mention relationships. Fandom networks are then transformed into fandom vectors through network embedding. Clustering the fandom vectors identified three types of K-pop fandoms: Influencer Followers, Interactive Coworkers, and Democratic Citizens. We explore the distinctive characteristics of each fandom type using social network metrics, enabling the development of customized strategies for engaging with each type of fandom. The findings and managerial implications of this study can also be applied to other media and entertainment industries.