EDNet: Attention-Based Multimodal Representation for Classification of Twitter Users Related to Eating Disorders

TY - GEN

T1 - EDNet

T2 - 2023 World Wide Web Conference, WWW 2023

AU - Abuhassan, Mohammad

AU - Anwar, Tarique

AU - Liu, Chengfei

AU - Jarman, Hannah K.

AU - Fuller-Tyszkiewicz, Matthew

N1 - Funding Information: This research is supported by the University of York, Australian Research Council (DP200103700), and Medical Research Future Fund (MRFF APP1179321). Publisher Copyright: © 2023 ACM.

PY - 2023/4/30

Y1 - 2023/4/30

N2 - Social media platforms provide rich data sources in several domains. In mental health, individuals experiencing an Eating Disorder (ED) are often hesitant to seek help through conventional healthcare services. However, many people seek help with diet and body image issues on social media. To better distinguish at-risk users who may need help for an ED from those who are simply commenting on ED in social environments, highly sophisticated approaches are required. Assessment of ED risks in such a situation can be done in various ways, and each has its own strengths and weaknesses. Hence, there is a need for and potential benefit of a more complex multimodal approach. To this end, we collect historical tweets, user biographies, and online behaviours of relevant users from Twitter, and generate a reasonably large labelled benchmark dataset. Thereafter, we develop an advanced multimodal deep learning model called EDNet using these data to identify the different types of users with ED engagement (e.g., potential ED sufferers, healthcare professionals, or communicators) and distinguish them from those not experiencing EDs on Twitter. EDNet consists of five deep neural network layers. With the help of its embedding, representation and behaviour modeling layers, it effectively learns the multimodalities of social media. In our experiments, EDNet consistently outperforms all the baseline techniques by significant margins. It achieves an accuracy of up to 94.32% and F1 score of up to 93.91% F1 score. To the best of our knowledge, this is the first such study to propose a multimodal approach for user-level classification according to their engagement with ED content on social media.

AB - Social media platforms provide rich data sources in several domains. In mental health, individuals experiencing an Eating Disorder (ED) are often hesitant to seek help through conventional healthcare services. However, many people seek help with diet and body image issues on social media. To better distinguish at-risk users who may need help for an ED from those who are simply commenting on ED in social environments, highly sophisticated approaches are required. Assessment of ED risks in such a situation can be done in various ways, and each has its own strengths and weaknesses. Hence, there is a need for and potential benefit of a more complex multimodal approach. To this end, we collect historical tweets, user biographies, and online behaviours of relevant users from Twitter, and generate a reasonably large labelled benchmark dataset. Thereafter, we develop an advanced multimodal deep learning model called EDNet using these data to identify the different types of users with ED engagement (e.g., potential ED sufferers, healthcare professionals, or communicators) and distinguish them from those not experiencing EDs on Twitter. EDNet consists of five deep neural network layers. With the help of its embedding, representation and behaviour modeling layers, it effectively learns the multimodalities of social media. In our experiments, EDNet consistently outperforms all the baseline techniques by significant margins. It achieves an accuracy of up to 94.32% and F1 score of up to 93.91% F1 score. To the best of our knowledge, this is the first such study to propose a multimodal approach for user-level classification according to their engagement with ED content on social media.

KW - Artificial intelligence

KW - Eating disorders

KW - Mental health

KW - Multimodal deep learning

KW - Online social media

UR - http://www.scopus.com/inward/record.url?scp=85159273968&partnerID=8YFLogxK

U2 - 10.1145/3543507.3583863

DO - 10.1145/3543507.3583863

M3 - Conference contribution

AN - SCOPUS:85159273968

T3 - ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023

SP - 4065

EP - 4074

BT - ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023

PB - Association for Computing Machinery, Inc

Y2 - 30 April 2023 through 4 May 2023

ER -