A Psychologically Informed Part-of-Speech Analysis of Depression in Social Media

Abstract

In this work, we provide an extensive part-ofspeech analysis of the discourse of social media users with depression. Research in psychology revealed that depressed users tend to be self-focused, more preoccupied with themselves and ruminate more about their lives and emotions. Our work aims to make use of largescale datasets and computational methods for a quantitative exploration of discourse. We use the publicly available depression dataset from the Early Risk Prediction on the Internet Workshop (eRisk) 2018 and extract partof-speech features and several indices based on them. Our results reveal statistically significant differences between the depressed and non-depressed individuals confirming findings from the existing psychology literature. Our work provides insights regarding the way in which depressed individuals are expressing themselves on social media platforms, allowing for better-informed computational models to help monitor and prevent mental illnesses.

Life is not Always Depressing: Exploring the Happy Moments of People Diagnosed with Depression

“…Events related to positive experiences with loved ones are more noteworthy in a depressed subject's life, more so than in the case of control subjects. Moreover, introspective discourse is more prevalent in depressed individuals (i.e., SELF category), which is in accordance with other works (Bucur et al, 2021b;Rude et al, 2004). Happy moments related to sleep are also more prevalent than those of the control category -poor sleep quality being one of the depression symptoms (O'Leary et al, 2017).…”

Tracking Mental Health Risks and Coping Strategies in Healthcare Workers’ Online Conversations Across the COVID-19 Pandemic

“…First-person singular pronouns (e.g., I, me, my) decreased across the pandemic, a pattern that was starker for nurses than for doctors. "I"-words typically indicate vulnerability to psychological distress (i.e., neuroticism or trait negative affectivity; Tackman et al 2019) and mental health concerns related to affect dysregulation, including depression (Bucur et al, 2021;Holtzman et al, 2017), anxiety (Brockmeyer et al, 2015;Shen and Rudzicz, 2017), eating disorders (Coppersmith et al, 2015a), and suicidality (Coppersmith et al, 2015b;Stirman and Pennebaker, 2001). Although the results were inconsistent with general psychological distress, a pattern of decreasing first-person singular pronoun usage is consistent with using self-distancing as a self-regulation strategy during periods of chronic stress.…”

An End-to-End Set Transformer for User-Level Classification of Depression and Gambling Disorder

“…Depression Depression detection from social media data is an interdisciplinary topic, and efforts have been made by researchers from both NLP and Psychology to detect different markers of depression found in the online discourse of individuals. Some depression cues found in language are: greater use of the first-person singular pronouns "I" [12], lesser use of first-person plural "we" [13], increased use of negative or absolutist terms (e.g., "never", "forever") [14], greater use of verbs at past tense [15].…”