Detection of Suicidal Ideation on Social Media: Multimodal, Relational, and Behavioral Analysis

Abstract

<jats:sec> <jats:title>Background</jats:title> <jats:p>Suicide risk assessment usually involves an interaction between doctors and patients. However, a significant number of people with mental disorders receive no treatment for their condition due to the limited access to mental health care facilities; the reduced availability of clinicians; the lack of awareness; and stigma, neglect, and discrimination surrounding mental disorders. In contrast, internet access and social media usage have increased significantly, providing experts and patients with a means of communication that may contribute to the development of methods to detect mental health issues among social media users.</jats:p> </jats:sec> <jats:sec> <jats:title>Objective</jats:title> <jats:p>This paper aimed to describe an approach for the suicide risk assessment of Spanish-speaking users on social media. We aimed to explore behavioral, relational, and multimodal data extracted from multiple social platforms and develop machine learning models to detect users at risk.</jats:p> </jats:sec> <jats:sec> <jats:title>Methods</jats:title> <jats:p>We characterized users based on their writings, posting patterns, relations with other users, and images posted. We also evaluated statistical and deep learning approaches to handle multimodal data for the detection of users with signs of suicidal ideation (suicidal ideation risk group). Our methods were evaluated over a dataset of 252 users annotated by clinicians. To evaluate the performance of our models, we distinguished 2 control groups: users who make use of suicide-related vocabulary (focused control group) and generic random users (generic control group).</jats:p> </jats:sec> <jats:sec> <jats:title>Results</jats:title> <jats:p>We identified significant statistical differences between the textual and behavioral attributes of each of the control groups compared with the suicidal ideation risk group. At a 95% CI, when comparing the suicidal ideation risk group and the focused control group, the number of friends (P=.04) and median tweet length (P=.04) were significantly different. The median number of friends for a focused control user (median 578.5) was higher than that for a user at risk (median 372.0). Similarly, the median tweet length was higher for focused control users, with 16 words against 13 words of suicidal ideation risk users. Our findings also show that the combination of textual, visual, relational, and behavioral data outperforms the accuracy of using each modality separately. We defined text-based baseline models based on bag of words and word embeddings, which were outperformed by our models, obtaining an increase in accuracy of up to 8% when distinguishing users at risk from both types of control users.</jats:p> </jats:sec> <jats:sec> <jats:title>Conclusions</jats:title> <jats:p>The types of attributes analyzed are significant for detecting users at risk, and their combination outperforms the results provided by generic, exclusively text-based baseline models. After evaluating the contribution of image-based predictive models, we believe that our results can be improved by enhancing the models based on textual and relational features. These methods can be extended and applied to different use cases related to other mental disorders.</jats:p> </jats:sec>

Evaluation of Korean-Language COVID-19–Related Medical Information on YouTube: Cross-Sectional Infodemiology Study

“…Demand-based infoveillance studies using internet search queries have primarily focused on predicting infectious disease outbreaks, such as Zika, influenza, dengue, and the measles virus [ 17 - 19 ]. Other studies analyzed user’s behavior on social media and proposed a model that was based on machine learning for the early detection of depression and suicidal risk [ 20 , 21 ].…”

Patterns of Routes of Administration and Drug Tampering for Nonmedical Opioid Consumption: Data Mining and Content Analysis of Reddit Discussions (Preprint)

“…Given the sensitive nature of the information shared, including users' vulnerabilities and personal information, privacy and ethical considerations are paramount. In this work, we followed the guidelines and directives in Eysenbach and Till [77], which describe recommendations to ethically conduct medical research with user-generated online data, and we relied on the vast experience of research works dealing with sensitive data gathered on social media [47,[78][79][80][81]. The researchers had no interactions with the users and have no interest in harming any, and the analyses were performed and reported in the spirit of knowledge, prevention, and harm reduction.…”

The ProfNER shared task on automatic recognition of occupation mentions in social media: systems, evaluation, guidelines, embeddings and corpora

“…The Spanish-speaking community on Twitter is large, exceeding 30 million (Tankovska, 2019), which motivated the implementation of text mining efforts for health-related applications, in particular on drug-related effects (Segura-Bedmar et al, 2015, 2014Ramírez-Cifuentes et al, 2020).…”