Abstract / Description of output
When studying how mental illness may be reflected in people’s social media use, content not written by the users is often ignored, because it might not reflect their own emotions. In this paper, we examine whether the mood of quotes posted on Facebook is affected by underlying symptoms of depression. We extracted quotes and song lyrics from the feeds of 781 Facebook users from the MyPersonality database who had also completed the CES-D depression scale. We found that participants with elevated depressive symptoms tend to post more song lyrics, especially lyrics with neutral or mixed sentiment. By analysing the topics of those lyrics, we found they center around overwhelming emotions, self-empowerment and retrospection of romantic relationships. Our findings suggest removing quotes, especially lyrics, might eliminate content that reflects users’ mental health conditions.
Original language | English |
---|---|
Title of host publication | SocInfo 2020: Social Informatics |
Publisher | Springer |
Pages | 58-66 |
Number of pages | 9 |
ISBN (Electronic) | 978-3-030-60975-7 |
ISBN (Print) | 978-3-030-60974-0 |
DOIs | |
Publication status | Published - 7 Oct 2020 |
Event | International Conference on Social Informatics - Pisa, Italy Duration: 6 Oct 2020 → 9 Oct 2020 https://kdd.isti.cnr.it/socinfo2020/ |
Publication series
Name | Lecture Notes in Computer Science |
---|---|
Publisher | Springer, Cham |
Volume | 12467 |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Conference
Conference | International Conference on Social Informatics |
---|---|
Abbreviated title | SocInfo 2020 |
Country/Territory | Italy |
City | Pisa |
Period | 6/10/20 → 9/10/20 |
Internet address |
Keywords / Materials (for Non-textual outputs)
- Social media
- Quotes
- Lyrics
- Depression