TY - GEN
T1 - WWBP-SQT-lite
T2 - 8th Workshop on Computational Linguistics and Clinical Psychology, CLPsych 2022
AU - Ganesan, Adithya V.
AU - Varadarajan, Vasudha
AU - Mittal, Juhi
AU - Subrahamanya, Shashanka
AU - Matero, Matthew
AU - Soni, Nikita
AU - Guntuku, Sharath Chandra
AU - Eichstaedt, Johannes C.
AU - Schwartz, H. Andrew
N1 - Publisher Copyright:
© 2022 Association for Computational Linguistics.
PY - 2022
Y1 - 2022
N2 - Psychological states unfold dynamically; to understand and measure mental health at scale we need to detect and measure these changes from sequences of online posts. We evaluate two approaches to capturing psychological changes in text: the first relies on computing the difference between the embedding of a message with the one that precedes it, the second relies on a "human-aware" multi-level recurrent transformer (HaRT). The mood changes of timeline posts of users were annotated into three classes, ‘ordinary,’ ‘switching’ (positive to negative or vice versa) and ‘escalations’ (increasing in intensity). For classifying these mood changes, the difference-between-embeddings technique – applied to RoBERTa embeddings – showed the highest overall F1 score (0.61) across the three different classes on the test set. The technique particularly outperformed the HaRT transformer (and other baselines) in the detection of switches (F1 = .33) and escalations (F1 = .61). Consistent with the literature, the language use patterns associated with mental-health related constructs in prior work (including depression, stress, anger and anxiety) predicted both mood switches and escalations.
AB - Psychological states unfold dynamically; to understand and measure mental health at scale we need to detect and measure these changes from sequences of online posts. We evaluate two approaches to capturing psychological changes in text: the first relies on computing the difference between the embedding of a message with the one that precedes it, the second relies on a "human-aware" multi-level recurrent transformer (HaRT). The mood changes of timeline posts of users were annotated into three classes, ‘ordinary,’ ‘switching’ (positive to negative or vice versa) and ‘escalations’ (increasing in intensity). For classifying these mood changes, the difference-between-embeddings technique – applied to RoBERTa embeddings – showed the highest overall F1 score (0.61) across the three different classes on the test set. The technique particularly outperformed the HaRT transformer (and other baselines) in the detection of switches (F1 = .33) and escalations (F1 = .61). Consistent with the literature, the language use patterns associated with mental-health related constructs in prior work (including depression, stress, anger and anxiety) predicted both mood switches and escalations.
UR - https://www.scopus.com/pages/publications/85137993151
M3 - Conference contribution
AN - SCOPUS:85137993151
T3 - CLPsych 2022 - 8th Workshop on Computational Linguistics and Clinical Psychology, Proceedings
SP - 251
EP - 258
BT - CLPsych 2022 - 8th Workshop on Computational Linguistics and Clinical Psychology, Proceedings
A2 - Zirikly, Ayah
A2 - Atzil-Slonim, Dana
A2 - Liakata, Maria
A2 - Bedrick, Steven
A2 - Desmet, Bart
A2 - Ireland, Molly
A2 - Lee, Andrew
A2 - MacAvaney, Sean
A2 - Purver, Matthew
A2 - Resnik, Rebecca
A2 - Yates, Andrew
PB - Association for Computational Linguistics (ACL)
Y2 - 15 July 2022
ER -