Summary
Computational mental health research faces challenges on decentralized social media because data and control are spread across different layers. The authors introduce a system that combines automated labeling, human checking, and modern language models to identify posts expressing suicidal thoughts or other mental health disclosures on Bluesky. They created large datasets labeled this way and found that model accuracy varies depending on how tasks are defined and validated. Human review showed different types of mistakes made by automatic labels, which points to the need for careful design when using decentralized platforms for mental health analysis.
What this means in practice
- •For mental health platform developers: Create reliable tools to detect suicidal and mental health content on decentralized social media to support user safety monitoring.
- •For social media data teams: Build validated mental health datasets from decentralized platform data streams to benchmark and improve content detection models.
Abstract
Decentralized social media platforms create new opportunities and challenges for computational mental health research because data access, moderation, labeling, and deployment responsibilities are distributed across multiple technical and governance layers. This paper presents a validation-aware weak-supervision system for constructing and evaluating suicidal ideation (SI) and broader mental health (MH) disclosure benchmarks on Bluesky, a decentralized social media platform built on the AT Protocol. The system integrates public firehose collection, task-specific lexicon filtering, Llama-3-8B-assisted binary annotation, human-adjudicated validation subsets, and transformer-based model benchmarking. Using this pipeline, we construct two task-specific corpora containing 8,346 SI-labeled posts and 9,988 MH-labeled posts. The evaluation shows that model performance depends strongly on both task definition and validation protocol. BERT+LSTM achieves the highest SI stratified cross-validation F1-score, RoBERTa achieves the strongest SI holdout F1-score, and DistilRoBERTa achieves the best MH cross-validation F1-score. Human validation reveals different weak-label failure modes across tasks, with SI labels dominated by false negatives and MH labels dominated by false positives. These findings show that decentralized social media can support reproducible mental health benchmarking, but only when system design, label provenance, validation strategy, and deployment constraints are evaluated together.