Skip to Main Content
Hundreds of thousands of people from different physical locations can join virtual worlds whenever they want to interact with each other and create objects in a computer-based simulated environment, typically in an interactive 3D format. The rich social media data generated in virtual worlds has important implications for business, education, social science, and society at large. However, how to collect such social media data is a relatively unexplored topic and remains an issue. In this study, we developed an integrated framework combining both bot-based and spider-based approaches to collect avatar behavioral and profile data. Following the proposed framework, we conducted a case study to examine avatar action differences between the two genders as well as between young-aged and old-aged avatars. The results of the case study indicated that in general male avatars tended to perform more active physical actions than female avatars, and young-aged avatars tended to perform more active physical actions than old-aged avatars.