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

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 4186–4192 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 4186 Life is not Always Depressing: Exploring the Happy Moments of People Diagnosed with Depression Ana-Maria Bucur1,2, Adrian Cosma3, Liviu P. Dinu1 1 University of Bucharest, Romania 2 Universitat Polit`ecnica de Val`encia, Spain 3 University Politehnica of Bucharest, Romania ana-maria.bucur@drd.unibuc.ro, cosma.i.adrian@gmail.com, ldinu@fmi.unibuc.ro Abstract In this work, we explore the relationship between depression and manifestations of happiness in social media. While the majority of works surrounding depression focus on symptoms, psychological research shows that there is a strong link between seeking happiness and being diagnosed with depression. We make use of Positive-Unlabeled learning paradigm to automatically extract happy moments from social media posts of both controls and users diagnosed with depression, and qualitatively analyze them with linguistic tools such as LIWC and keyness information. We show that the life of depressed individuals is not always bleak, with positive events related to friends and family being more noteworthy to their lives compared to the more mundane happy events reported by control users. Keywords: happiness, depression, social media, positive-unlabeled learning, LIWC, BERT 1. Introduction ”I have the choice of being constantly active and happy or introspectively passive and sad. Or I can go mad by ricocheting in between”, wrote the famous poet Sylvia Plath in her personal diaries (Plath and Hughes, 2000). It is now known that Sylvia Plath was clinically de- pressed most of her life, before tragically committing suicide. Depression is the modern ailment of our times. It is reported that more than 264 million people of all ages suffer from this mental illness 1. More than 700 thousand people take their own life every year, and even more people attempt suicide 2. Thankfully, mod- ern psychology has a fair understanding of depression, its onset, its symptoms and potential treatments, more recently aided by computational methods (Orabi et al., 2018; Burdisso et al., 2019; Uban and Rosso, 2020; Inkpen et al., 2021; Bucur and Dinu, 2020; Bucur et al., 2021a). However, research in automatic understanding of manifestations of moments of happiness is lacking. How much do we understand about the happy moments of depressed individuals? In this preliminary work, we aim to bridge this gap and develop a computational method for extracting and analyzing happy moments from a large corpus of social media text. Indeed, promoting happiness is an important aspect of our lives, and is one of the main ways in which we can prevent disease and live a longer, healthier life. While our happiness is determined largely by genetic factors (Diener et al., 1999), it can be influenced in a significant manner through our behavior: long-term re- lationships with friends and family, meaningful work and self-care. Paradoxically, Ford et al. (2014) showed that individuals in cultures that highly value happiness 1https://www.who.int/news-room/ fact-sheets/detail/depression 2https://www.who.int/news-room/ fact-sheets/detail/suicide are less likely to attain it. Moreover, the authors show that highly valuing happiness is a key indicator …

Content truncated to 500 words