Araştırma Makalesi

Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM

Cilt: 13 Sayı: 3 7 Eylül 2026
PDF İndir
EN TR

Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM

Öz

Mental health risk detection from user-generated social media text has become increasingly important as psychiatric conditions such as depression, anxiety, and suicidal ideation continue to rise. However, most existing computational studies operationalize the problem using single-label datasets, implicitly assuming mutually exclusive conditions and thereby under-modeling clinically prevalent comorbidity. To better align modeling assumptions with real-world mental health phenomena, we formulate social-media-based risk identification as a multi-label text classification problem, enabling the simultaneous prediction of co-existing risks within a single post. We construct a hybrid corpus by integrating AIMH/SWMH and the Sentiment Analysis for Mental Health dataset, and by adding neutral/positive samples from Sentiment140 to strengthen healthy-content representation. Multi-label annotations are generated via Meta Llama-3-70B-Instruct using deterministic zero-shot prompting (temperature=0); a manual audit of 1,000 randomly sampled instances yields 94% agreement with the LLM-generated labels. We then benchmark multiple transformer-based encoders under a unified multi-label training protocol and report macro/micro F1 performance, highlighting the feasibility of transformer-based multi-label learning for modeling overlapping mental health risks at scale in social media.

Anahtar Kelimeler

Kaynakça

  1. [1] World Health Organization, "Mental Health Atlas 2024", World Health Organization, Geneva, (2025).
  2. [2] Canbay P., "Predicting discriminative personality profile of haters from digital texts", Knowledge-Based Systems, 2024, 287: 111460.
  3. [3] Zhang T., Schoene A.M., Ji S., Ananiadou S., "Natural language processing applied to mental illness detection: A narrative review", NPJ Digital Medicine, 2022, 5(1): 46.
  4. [4] Montejo-Raez A., Molina-Gonzalez M.D., Jimenez-Zafra S.M., Garcia-Cumbreras M.A., Garcia-Lopez L.J., "A survey on detecting mental disorders with natural language processing: Literature review, trends and challenges", Computer Science Review, 2024, 53: 100654.
  5. [5] Platona R.I., Căiţă G.A., Voiţă-Mekeres F., Peia A.O., Enătescu R.V., "The impact of psychiatric comorbidities associated with depression: A literature review", Medicine and Pharmacy Reports, 2024, 97(2): 143.
  6. [6] Ji S., Li X., Huang Z., Cambria E., "Suicidal ideation and mental disorder detection with attentive relation networks", Neural Computing and Applications, 2021, 34(13): 10309-10319.
  7. [7] Ezerceli Ö., Dehkharghani R., "Mental disorder and suicidal ideation detection from social media using deep neural networks", Journal of Computational Social Science, 2024, 7(3): 2277-2307.
  8. [8] Go A., Bhayani R., Huang L., "Twitter sentiment classification using distant supervision", CS224N Project Report, Stanford University, 2009, 1: 1-12.

Ayrıntılar

Birincil Dil

İngilizce

Konular

Mühendislik Tasarımı

Bölüm

Araştırma Makalesi

Yayımlanma Tarihi

7 Eylül 2026

Gönderilme Tarihi

10 Şubat 2026

Kabul Tarihi

14 Nisan 2026

Yayımlandığı Sayı

Yıl 2026 Cilt: 13 Sayı: 3

Kaynak Göster

APA
Özer, Z., Kenger, E., Gözükara, G., & Canbay, P. (2026). Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM. El-Cezeri, 13(3), 383-392. https://doi.org/10.31202/ecjse.1885302
AMA
1.Özer Z, Kenger E, Gözükara G, Canbay P. Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM. ECJSE. 2026;13(3):383-392. doi:10.31202/ecjse.1885302
Chicago
Özer, Zehra, Emre Kenger, Görkem Gözükara, ve Pelin Canbay. 2026. “Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM”. El-Cezeri 13 (3): 383-92. https://doi.org/10.31202/ecjse.1885302.
EndNote
Özer Z, Kenger E, Gözükara G, Canbay P (01 Eylül 2026) Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM. El-Cezeri 13 3 383–392.
IEEE
[1]Z. Özer, E. Kenger, G. Gözükara, ve P. Canbay, “Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM”, ECJSE, c. 13, sy 3, ss. 383–392, Eyl. 2026, doi: 10.31202/ecjse.1885302.
ISNAD
Özer, Zehra - Kenger, Emre - Gözükara, Görkem - Canbay, Pelin. “Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM”. El-Cezeri 13/3 (01 Eylül 2026): 383-392. https://doi.org/10.31202/ecjse.1885302.
JAMA
1.Özer Z, Kenger E, Gözükara G, Canbay P. Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM. ECJSE. 2026;13:383–392.
MLA
Özer, Zehra, vd. “Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM”. El-Cezeri, c. 13, sy 3, Eylül 2026, ss. 383-92, doi:10.31202/ecjse.1885302.
Vancouver
1.Zehra Özer, Emre Kenger, Görkem Gözükara, Pelin Canbay. Modeling Co-Existing Mental Health Risks in Social Media via Multi-Label Learning and LLM. ECJSE. 01 Eylül 2026;13(3):383-92. doi:10.31202/ecjse.1885302