How are you doing? : emotions and personality in Facebook

User generated content on social media sites is a rich source of information about latent variables of their users. Proper mining of this content provides a shortcut to emotion and personality detection of users without filling out questionnaires. This in turn increases the application potential of personalized services that rely on the knowledge of such latent variables. In this paper we contribute to this emerging domain by studying the relation between emotions expressed in approximately 1 million Facebook (FB) status updates and the users' age, gender and personality. Additionally, we investigate the relations between emotion expression and the time when the status updates were posted. In particular, we find that female users are more emotional in their status posts than male users. In addition, we find a relation between age and sharing of emotions. Older FB users share their feelings more often than young users. In terms of seasons, people post about emotions less frequently in summer. On the other hand, December is a time when people are more likely to share their positive feelings with their friends. We also examine the relation between users' personality and their posts. We find that users who have an open personality express their emotions more frequently, while neurotic users are more reserved to share their feelings