Multi-task learning to detect suicide ideation and mental disorders among social media users

Abstract

Mental disorders and suicide are considered global health problems faced by many countries worldwide. Even though advancements have been made to improve mental wellbeing through research, there is room for improvement. Using Artificial Intelligence to early detect individuals susceptible to mental illness and suicide ideation based on their social media postings is one way to start. This research investigates the effectiveness of using a shared representation to automatically extract features between the two different yet related tasks of mental illness and suicide ideation detection using data in parallel from social media platforms with different distributions. In addition to discovering the shared features between users with suicidal thoughts and users who self-declared a single mental disorder, we further investigate the impact of comorbidity on suicide ideation and use two datasets during inference to test the generalizability of the trained models and provide satisfactory evidence to validate the increased predictive accurateness of suicide risk when using data from users diagnosed with multiple mental disorders compared to a single mental disorder for the mental illness detection task. Our results also demonstrate different mental disorders' impact on suicidal risk and discover a noticeable impact when using data from users diagnosed with Post-Traumatic Stress Disorder. We use multi-task learning (MTL) with soft and hard parameter sharing to produce state-of-the-art results for detecting users with suicide ideation who require urgent attention. We further improve the predictability of the proposed model by demonstrating the effectiveness of cross-platform knowledge sharing and predefined auxiliary inputs.

AI for Analyzing Mental Health Disorders Among Social Media Users: Quarter-Century Narrative Review of Progress and Challenges (Preprint)

“…SMHD, meanwhile, was born out of a desire for data sets covering a broad range of mental health disorders. It provided a platform for development of methods concerning not only depression [63,64], but also suicidal ideation [65], schizophrenia [66], and even multi-class experimental setups involving combinations of anxiety, eating disorders, ADHD, bipolar disorder, and PTSD [67,68,69,70]. It was also intended that a wider range of higher positive predictive value patterns be used to collect a greater volume of diagnosed users.…”

MM-EMOG: Multi-Label Emotion Graph Representation for Mental Health Classification on Social Media

“…Recent studies have focused on incorporating more social media components to capture as much available contextual information as possible. Among these are historical posts [4,[9][10][11][12][13][14][15][16][17][18], conversation trees [19], social and interaction graphs [4,10,12,17,20], user and post metadata information [10,11], and images [10]. While more contextual sources may be ideal for assessing an individual's mental health state, access to these data has become increasingly restrictive due to heightened data privacy concerns.…”

Enhancing Mental Health Condition Detection on Social Media through Multi-Task Learning

“…Their approach utilized a straightforward feedforward neural network designed to manage multiple related tasks by sharing a single hidden layer among them. Subsequently, Buddhitha and Inkpen 12 continued exploration of multitask mental disorder detection employing a multi-channel convolutional neural network (CNN) 13 within a multi-task learning framework, incorporating specific auxiliary inputs to identify individuals with post-traumatic stress disorder (PTSD) and depression. Despite these methods showing potential in the collective modeling of mental disorders, they struggled to achieve good performance.…”