A cognitive categorization-based approach for understanding identity representation online

[摘要] Computationally representing social identities using social networking profiles traditionally involves the reduction of identities to fit into simplistic categories such as ;;friends.;; In contrast, this thesis proposes that the data structures underlying user identities can be algorithmically processed and interpreted in ways that assist in understanding more nuanced aspects of identity such as ;;subculture;; or;;personality;; Building upon an interdisciplinary computational identity model developed by Fox Harrell in his NSF-supported Advanced Identity Representation Project, this thesis proposes an algorithm based on theories of cognitive categorization[6, 7] to reveal implicit categories in computational identity systems. The algorithm has been applied to social networking site Facebook and a suite of graphical user interfaces was developed to enable users to explore individual and group identities. In a qualitative study, we found that most of the generated categories coherently represented social groups and would be useful for applications such as expressing the groups;; collective identities.

[发布机构] Massachusetts Institute of Technology