Multi-task learning to detect suicide ideation and mental disorders among social media users
Bibliographic record
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.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.003 | 0.004 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".