A friend in need: quantifying the role of responses to expressions of distress on social media
Tomer Sagi - Department of Information Systems
Reout Arbel# and Keren Segal - Faculty of Education
Machine Learning
Deep learning
Neural Networks
Psychology
Database Collection Grant 2021
A growing number of people are using social media daily. It has become a venue for people to express their feelings and thoughts, even when in distress and seeking support. However, it is unclear how much of this support is received and whether it is effective. Much research has been done on automating the ability to identify people in distress in general, or specifically those at risk of committing suicide, many of which are teenagers and young adults who use social media frequently. However, there are no automated tools to identify whether a person is receiving a supportive response from his/her peers or not.
In exploratory work, Keren Segal is examining how responses to individuals expressing distress affect their future expressions of stress. As a first step, we have obtained a dataset of more than 600 million Tweets and Re-tweets in English belonging to 10M Twitter users that were collected during 2015 in the USA to use as our main dataset in this work. From this dataset, we are using an automated machine-learning model [1] trained on the University of Maryland (UMD) suicidal ideation dataset [2] to retrieve suicidal ideation tweets and their subsequent responses. The response will be paired with the original post and presented as a task to be tagged according to the type (or lack thereof) of support offered. The results will be used to explore how social support affects the subsequent online behavior of distressed individuals. We are currently in the process of refining the automated suicidal ideation detection method to work on a tweet-level to identify the suicidal tweets themselves since the original training data we received from UMD only identifies suicidal users.
Keren is pursuing a Master’s degree in Knowledge and Information Management under the supervision of Dr. Tomer Sagi of the Department of Information Systems and Dr. Reout Arbel of the Department of Counseling and Human Development at the University of Haifa.
Keren’s work is generously supported by the Haifa University Data Science Research Center (DSRC).
[1] S. Ji, C. P. Yu, S.-f. Fung, S. Pan, and G. Long. Supervised learning for suicidal ideation detection in online user content. Complexity, 2018.
[2] H.-C. Shing, S. Nair, A. Zirikly, M. Friedenberg, H. Daum´e III, and P. Resnik. Expert, crowdsourced, and machine assessment of suicide risk via online postings. In Proceedings of the Fifth Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, pages 25–36, 2018.
A growing number of people are using social media daily. It has become a venue for people to express their feelings and thoughts, even when in distress and seeking support. However, it is unclear how much of this support is received and whether it is effective. Much research has been done on automating the ability to identify people in distress in general, or specifically those at risk of committing suicide, many of which are teenagers and young adults who use social media frequently. However, there are no automated tools to identify whether a person is receiving a supportive response from his/her peers or not.
In exploratory work, Keren Segal is examining how responses to individuals expressing distress affect their future expressions of stress. As a first step, we have obtained a dataset of more than 600 million Tweets and Re-tweets in English belonging to 10M Twitter users that were collected during 2015 in the USA to use as our main dataset in this work. From this dataset, we are using an automated machine-learning model [1] trained on the University of Maryland (UMD) suicidal ideation dataset [2] to retrieve suicidal ideation tweets and their subsequent responses. The response will be paired with the original post and presented as a task to be tagged according to the type (or lack thereof) of support offered. The results will be used to explore how social support affects the subsequent online behavior of distressed individuals. We are currently in the process of refining the automated suicidal ideation detection method to work on a tweet-level to identify the suicidal tweets themselves since the original training data we received from UMD only identifies suicidal users.
Keren is pursuing a Master’s degree in Knowledge and Information Management under the supervision of Dr. Tomer Sagi of the Department of Information Systems and Dr. Reout Arbel of the Department of Counseling and Human Development at the University of Haifa.
Keren’s work is generously supported by the Haifa University Data Science Research Center (DSRC).
[1] S. Ji, C. P. Yu, S.-f. Fung, S. Pan, and G. Long. Supervised learning for suicidal ideation detection in online user content. Complexity, 2018.
[2] H.-C. Shing, S. Nair, A. Zirikly, M. Friedenberg, H. Daum´e III, and P. Resnik. Expert, crowdsourced, and machine assessment of suicide risk via online postings. In Proceedings of the Fifth Workshop on Computational Linguistics and Clinical Psychology: From Keyboard to Clinic, pages 25–36, 2018.