Understanding Cross-Platform Links in User Profiles: A Data-Driven Analysis on Mastodon
摘要
Online social networks (OSNs) are extremely popular in people’s daily life. Mastodon, as a representative decentralized OSN, have attracted attentions from both academia and industry in recent years. Mastodon users are allowed to add external links in their profiles, pointing to profiles on other OSNs, personal websites, or project repositories. These links serve as signals of identity and presence beyond the local instance. However, the characteristics of users who include such links, and how their behaviors differ within decentralized OSNs, remain underexplored. This study examines the demographic and behavioral characteristics of Mastodon users who include external links in their profiles. Based on a dataset of over two million users, we find that more than ten percent users link to external platforms. These users tend to be more active and better connected within the network. However, further causal analysis reveals that platform discoverability, rather than social metrics such as follower count or posting frequency, is the simple behavioral driver of linking behavior. These findings suggest that profile linking in decentralized OSNs reflects intentional identity signaling shaped by visibility affordances, rather than being a passive outcome of network activity.