Suicide Risk Prediction by Tracking Self-Harm Aspects in Tweets: NUS-IDS at the CLPsych 2021 Shared Task

Sujatha Das Gollapalli, Guilherme Augusto Zagatti, See-Kiong Ng


Abstract
We describe our system for identifying users at-risk for suicide based on their tweets developed for the CLPsych 2021 Shared Task. Based on research in mental health studies linking self-harm tendencies with suicide, in our system, we attempt to characterize self-harm aspects expressed in user tweets over a period of time. To this end, we design SHTM, a Self-Harm Topic Model that combines Latent Dirichlet Allocation with a self-harm dictionary for modeling daily tweets of users. Next, differences in moods and topics over time are captured as features to train a deep learning model for suicide prediction.
Anthology ID:
2021.clpsych-1.10
Volume:
Proceedings of the Seventh Workshop on Computational Linguistics and Clinical Psychology: Improving Access
Month:
June
Year:
2021
Address:
Online
Editors:
Nazli Goharian, Philip Resnik, Andrew Yates, Molly Ireland, Kate Niederhoffer, Rebecca Resnik
Venue:
CLPsych
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
93–98
Language:
URL:
https://aclanthology.org/2021.clpsych-1.10
DOI:
10.18653/v1/2021.clpsych-1.10
Bibkey:
Cite (ACL):
Sujatha Das Gollapalli, Guilherme Augusto Zagatti, and See-Kiong Ng. 2021. Suicide Risk Prediction by Tracking Self-Harm Aspects in Tweets: NUS-IDS at the CLPsych 2021 Shared Task. In Proceedings of the Seventh Workshop on Computational Linguistics and Clinical Psychology: Improving Access, pages 93–98, Online. Association for Computational Linguistics.
Cite (Informal):
Suicide Risk Prediction by Tracking Self-Harm Aspects in Tweets: NUS-IDS at the CLPsych 2021 Shared Task (Gollapalli et al., CLPsych 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.clpsych-1.10.pdf
Optional supplementary code:
 2021.clpsych-1.10.OptionalSupplementaryCode.zip
Optional supplementary data:
 2021.clpsych-1.10.OptionalSupplementaryData.zip