Paper: I’m a Belieber: Social Roles via Self-identification and Conceptual Attributes

ACL ID P14-2030
Title I’m a Belieber: Social Roles via Self-identification and Conceptual Attributes
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2014
Authors

Motivated by work predicting coarse- grained author categories in social me- dia, such as gender or political preference, we explore whether Twitter contains infor- mation to support the prediction of fine- grained categories, or social roles. We find that the simple self-identification pat- tern ?I am a ? supports significantly richer classification than previously ex- plored, successfully retrieving a variety of fine-grained roles. For a given role (e.g., writer), we can further identify character- istic attributes using a simple possessive construction (e.g., writer?s ). Tweets that incorporate the attribute terms in first person possessives (my ) are confirmed to be an indicator that the author holds the associated social role.