Grief-Stricken in a Crowd: The Language of Bereavement and Distress in Social Media Jed R. Brubaker 1 , Funda Kivran-Swaine 2 , Lee Taber 1 , and Gillian R. Hayes 1 1 School of Information and Computer Sciences University of California, Irvine Irvine, CA, USA {jed.brubaker, ltaber, hayesg}@uci.edu 2 School of Communication and Information Rutgers University New Brunswick, NJ, USA funda@rutgers.edu Abstract People turn to social media to express their emotions surrounding major life events. Death of a loved one is one scenario in which people share their feelings in the semi- public space of social networking sites. In this paper, we present the results of a two-part investigation of grief and distress in the context of messages posted to the profiles of deceased MySpace users. We present coding system for identifying emotion distressed content, followed by a detailed analysis of language use that lays a foundation for natural language processing (NLP) tasks, such as automatic detection of bereavement-related distress. Our findings suggest that in addition to words bearing positive or negative sentiment, linguistic style can be an indicator of messages that demonstrate distress in the space of post- mortem social media content. These results contribute to research in computational linguistics by identifying linguistic features that can be used for automatic classification as well as to research on death and bereavement by enumerating attributes of distressed self- expression in a post-mortem context. Introduction Social media-based interactions are an increasingly important to many people’s social lives. Life’s significant events, from birthdays to marriage proposals, are now experienced in part via social network sites (SNS). This remains true of when those lives come to an end. Grieving a loved one’s death—once limited to the physical world— now extends to the social networking profiles of the deceased. Following a user’s death, friends often express their shock and grief on the deceased’s profile page (Brubaker & Vertesi 2010; Getty et al. 2011). Over time, survivors continue to return to social media and keep those who have passed close through continued updates on events and feelings that would have been shared with the deceased were they alive (Brubaker & Hayes 2011). Copyright © 2011, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. Analyzing content from sites such as MySpace provides insight into the role that social media plays in contemporary experiences of death, as well as the impact of death on these digital spaces. So far, much of the previous work in the area of post-mortem interactions in social media has been qualitative in approach, employing content and thematic analyses to understand grieving practices. However, quantitative analyses of the content and linguistic style of post-mortem comments are crucial in understanding practices during bereavement. Moreover, user-generated content posted via social media provides a unique opportunity to observe grieving practices in a naturalistic setting. Messages in SNS are in their intended social environments as opposed to interview or lab settings, where expressions of grief have historically been studied. This work focuses on extreme expressions of grief and mourning in SNS following the death of a friend or loved one. In the first part of this paper we present a codebook for identifying distressed comments based on a thorough review of the literature on grieving practices and bereavement. This coding scheme identifies comments exhibiting emotional distress as a potential indicator of unhealthy mourning and/or the need for additional support. Based on comments coded through this scheme, in the second part of this paper we present the results of a detailed linguistic analysis of comments posted to deceased MySpace profiles in order build explanatory models based on linguistic features that contrast comments exhibiting emotional distress with those that display more conventional funerary language. To this end, we examined 2213 post-mortem comments posted to the profiles of 652 MySpace users following their deaths. The detailed linguistic analysis of distressed comments presented here expands our current knowledge around the use of language in online grieving. Existing research in this space has predominantly focused on the benefits of such spaces that enable individuals to express their grief. A review of content, however, reveals that some of the