Monitoring Tweets for Depression to Detect At-risk Users

Zunaira Jamil, Diana Inkpen, Prasadith Buddhitha, Kenton White


Abstract
We propose an automated system that can identify at-risk users from their public social media activity, more specifically, from Twitter. The data that we collected is from the #BellLetsTalk campaign, which is a wide-reaching, multi-year program designed to break the silence around mental illness and support mental health across Canada. To achieve our goal, we trained a user-level classifier that can detect at-risk users that achieves a reasonable precision and recall. We also trained a tweet-level classifier that predicts if a tweet indicates depression. This task was much more difficult due to the imbalanced data. In the dataset that we labeled, we came across 5% depression tweets and 95% non-depression tweets. To handle this class imbalance, we used undersampling methods. The resulting classifier had high recall, but low precision. Therefore, we only use this classifier to compute the estimated percentage of depressed tweets and to add this value as a feature for the user-level classifier.
Anthology ID:
W17-3104
Volume:
Proceedings of the Fourth Workshop on Computational Linguistics and Clinical Psychology — From Linguistic Signal to Clinical Reality
Month:
August
Year:
2017
Address:
Vancouver, BC
Venues:
CLPsych | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
32–40
Language:
URL:
https://aclanthology.org/W17-3104
DOI:
10.18653/v1/W17-3104
Bibkey:
Cite (ACL):
Zunaira Jamil, Diana Inkpen, Prasadith Buddhitha, and Kenton White. 2017. Monitoring Tweets for Depression to Detect At-risk Users. In Proceedings of the Fourth Workshop on Computational Linguistics and Clinical Psychology — From Linguistic Signal to Clinical Reality, pages 32–40, Vancouver, BC. Association for Computational Linguistics.
Cite (Informal):
Monitoring Tweets for Depression to Detect At-risk Users (Jamil et al., 2017)
Copy Citation:
PDF:
https://aclanthology.org/W17-3104.pdf