Research map / Large language models / AI and society
AI in the Wild: A Meta-Analytic Evaluation of Depression Detection from Social Media Data
Xianglu Tang, Joyee W. Jin, Emily Ma, Xingyu Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use deep learning or hybrid architectures instead of traditional ML (ablation) | social-media depression detectors · 67 studies, adjusted meta-regression | accuracy / F1 / r: beta Accuracy 0.0498, F1 0.0472, r 0.0609 | Improved |
| add multimodal inputs (images, audio, graph) beyond text (ablation) | depression detectors · 67 studies | F1 / accuracy / r: text-only F1 -0.0738, accuracy -0.0453; r unaffected (p = .184) | Mixed |
| publication year 2024 vs 2022-23 | depression detectors · full corpus | Pearson r: +0.16 r (beta 0.158) | Improved |
| use clinical-grade labels (scales, clinician) instead of crowd tags | depression detectors · 67 studies | performance: t = -0.12, p < .001 | Hurt |
| ethical compliance (IRB, privacy safeguards) | depression detectors · 67 studies | performance: p > .05 | No clear effect |
| cross-platform testing instead of in-platform | depression detectors · 5 cross-platform studies | Pearson r / F1: -42.5 pp r, -4.5% F1 | Hurt |
| user-level severity estimation vs post-level binary task | depression detectors · task granularity | Pearson r: 0.75 -> 0.25 | Hurt |
Closest papers by meaning
- Mental-Perceiver: Audio-Textual Multi-Modal Learning for Estimating Mental Disorders (AAAI 2025 · similarity 0.85)
- Depression Detection via Capsule Networks with Contrastive Learning (AAAI 2024 · similarity 0.85)
- MoodAngels: A Retrieval-augmented Multi-agent Framework for Psychiatry Diagnosis (NeurIPS 2025 · similarity 0.85)
- MAMA-Memeia! Multi-Aspect Multi-Agent Collaboration for Depressive Symptoms Identification in Memes (AAAI 2026 · similarity 0.85)
- EmoNet-Face: An Expert-Annotated Benchmark for Synthetic Emotion Recognition (NeurIPS 2025 · similarity 0.85)
- Thousand Voices of Trauma: A Large-Scale Synthetic Dataset for Modeling Prolonged Exposure Therapy Conversations (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
