Pith. sign in

REVIEW 4 major objections 6 minor 35 references

Detecting anxiety and depression in dialogues: a multi-label and explainable approach

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A multi-label system detects anxiety and depression from chatbot dialogues with 90% accuracy, pairing LLM feature extraction with random forest classification.

desk verdict The 90% accuracy claim is invalid: the questionnaire responses used as labels are in the same session text the LLM feature extractor sees, so the model is partly reading the label source. read the letter →

arxiv 2412.17651 v1 pith:OIPLDKCR submitted 2024-12-23 cs.AI

Francisco de Arriba-Pérez , Silvia García-Méndez This is my paper · ORCID
classification cs.AI
keywords anxietydetectiondepressionmulti-labelclassificationlargelanguagemodelsexplainableAIchatbotdialoguesmentalhealthscreeningnaturalprocessing
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that anxiety and depression can be detected together, as a multi-label problem, from free-form conversations with a chatbot. The proposed system uses a large language model to score fourteen emotional and linguistic cues in each session, then feeds those scores plus their statistics over past sessions into standard machine-learning classifiers. On 2,186 real sessions from the Celia chatbot, the best classifier, Random Forest, reaches 89.43% accuracy with a 10.57% Hamming loss, outperforming the prior literature the authors compare against. If true, this would make early screening scalable and less reliant on rigid questionnaires, because the assessment happens naturally inside an ongoing dialogue and the model explains its decision in plain language.

What carries the argument

The machinery is a two-stage pipeline. First, a prompt-engineered LLM (gpt-4o-mini with temperature 0) reads each full dialogue session and outputs scores from 0.0 to 1.0 for 14 features covering the user's cognitive state (insecurity, loneliness, negative/positive emotion, sadness, anguish, health issues) and the discourse (catastrophic terms, emphasized terms, repeated concepts, interjections, negative adverbs, negative terms, polarity). Each feature is expanded with its average and three quartiles over a sliding window of the last 30 sessions, yielding 56 features; a tree-based impurity selector keeps about 39% of them. The selected features train Naive Bayes, Decision Tree, and Random Forest classifiers under a multi-class transformation of the two clinical labels, and the best model's prediction is turned into a natural-language explanation by a second prompt template that also fills a dashboard for caregivers and users.

What would settle it

Run a leave-one-user-out cross-validation on the same dataset: if accuracy on users never seen in training drops substantially below the reported 89.43%, the performance is partly user memorization rather than general detection. A second check is to re-label a sample of sessions with the same questionnaires at different times and measure how often the label changes for the same user's sessions; high instability would show the model is learning noisy targets.

Watch

Extended reading notes

Core claim

The central claim is that coupling an LLM feature-engineering stage with a traditional multi-label classifier yields accurate joint detection of anxiety and depression from real chatbot dialogues, with explainability preserved. The authors treat the task as four mutually exclusive categories—no condition, depression only, anxiety only, and both—using the multi-class transformation strategy on binary labels derived from Spanish versions of the Goldberg Anxiety and Depression Scales and the Yesavage Geriatric Depression Scale administered inside the chatbot every three months. They report that Random Forest attains 89.43% accuracy, 93.50% macro precision, 84.18% macro recall, and 84.72% exact-match ratio in the full-session scenario, and that these results beat the closest related multi-label baseline, which used activity-tracker data and binary-relevance classifiers.

Load-bearing premise

The evaluation assumes that the 2,186 sessions from 32 users are independent and that the labels, assigned every three months from questionnaires embedded in the chatbot, are correct and stable; if users' repeated sessions or label noise are not handled, the reported accuracy overstates how the system performs on independent individuals.

Editorial extensions

If this is right

  • If the 90%-level accuracy holds, a health system could deploy the pipeline as a pre-screening layer that flags users for formal assessment without making them fill out anxiety and depression questionnaires separately.
  • Because the LLM is used only for feature extraction, the final classification is done by an interpretable ML model, so the system avoids relying on the LLM as a black-box judge and still gives a confidence score.
  • The sliding-window statistics mean the model uses the user's history, not just one isolated chat, so a single atypical session should not by itself change the assessment.
  • The explainability dashboard could let clinicians see which of the fourteen cues drove the prediction, supporting accountability under AI-regulation expectations that users receive clear explanations.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not report user-level cross-validation; the natural next test is leave-one-user-out evaluation, because with only 32 users, standard cross-validation may mix sessions from the same user in both training and test sets.
  • The quarterly re-labeling creates a temporal dependency: the same user's sessions share a label for up to three months, so adjacent sessions are not independent, and a stricter evaluation would group sessions by labeling period.
  • The fourteen features are generic enough that the same prompt-template approach could be tested on other disorders or other languages, but that would require retraining and revalidating labels.
  • A practical extension the authors do not report is calibration: knowing when the confidence score is unreliable could matter more than average accuracy in a pre-screening tool that decides who gets referred.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes a multi-label classifier for anxiety and depression from free-form chatbot dialogues. It uses GPT-4o-mini to extract 14 psychological and linguistic features from each user session via a prompt, expands each feature with sliding-window statistics, selects features on a 10% cold-start subset, and trains Naive Bayes, decision tree, and random forest classifiers. Reported results in Table 4 claim up to 89.43% accuracy and 84.72% exact match for the random forest in scenario 1, which the authors state improves on prior literature. An explainability dashboard is also presented. The main technical claim is that an LLM-feature-based ML pipeline can jointly detect anxiety and depression from natural conversation.

Significance. If the central claim were supported, the paper would offer a practical combination of LLM feature extraction with interpretable multi-label classifiers and an explainability dashboard, addressing a real need for scalable mental-health screening. The use of a real chatbot dataset rather than social media, and the effort to provide clinician-facing explanations, are welcome. However, the reported 90% accuracy is not currently credible as evidence of dialogue-based detection, because the labels are derived from questionnaires embedded in the same session text that is fed to the LLM feature extractor. The evaluation also lacks a proper held-out test set, user-level splits, baselines, and error bars. The paper's contribution is therefore potentially useful but its headline result is unsupported.

major comments (4)
  1. [Sec. 3.1, Sec. 4.2, Listing 1] There is a structural label-leakage problem: the GADS and YGDS questionnaires are embedded in the conversation flow every three months, and the responses to those questionnaires are used as the target labels. The LLM feature extractor in Listing 1 receives the complete session text, which therefore includes the questionnaire items and the user's answers. Features such as "anguish" and "health issues" can be read directly from the user's questionnaire responses rather than inferred from free-form dialogue. The high accuracy in Table 4 is thus fully compatible with the model recognizing the label source inside the input, and it does not support the claim that the system detects anxiety and depression from free-form conversation. The authors need to re-run feature extraction after removing all questionnaire-related text from the session input, or otherwise demonstrate that the questionnaire content is not accessible to the feature extractor.
  2. [Sec. 4.4, Sec. 4.5, Table 4] The evaluation does not use a proper held-out test set. Feature selection is performed on 10% of the dataset (Sec. 4.4), and hyperparameters are optimized with GridSearchCV on the same 10% subset (Sec. 4.5). The paper does not state that the final models are evaluated on a disjoint test set, so the results in Table 4 likely reflect training-data performance or at least tuning leakage. A nested cross-validation or a fixed train/validation/test split with feature selection and hyperparameter tuning performed entirely within the training folds is required before the accuracy claims can be accepted.
  3. [Sec. 4.1, Sec. 4.2] The dataset consists of 2,186 sessions from only 32 users, with an average of 68 sessions per user, and labels are re-assigned every three months from the embedded questionnaires. The experiments do not group sessions by user or by time period. Random session-level splits will almost certainly place sessions from the same user in both training and test folds, and temporally adjacent sessions from the same user have strongly correlated labels and features. This can substantially inflate accuracy. The authors should use user-disjoint and temporally split evaluation, e.g., leaving out entire users or entire time periods, to estimate performance on independent individuals.
  4. [Sec. 4.5, Sec. 4.6, Table 4] The paper claims the results "improve those in the prior literature," but it provides no quantitative baselines on the same dataset or protocol. There is no comparison against simple text baselines such as bag-of-words with logistic regression, or against transformer or LLM classifiers, and no confidence intervals or significance tests are reported. Without these, the 90% accuracy figure cannot be interpreted. The explainability dashboard is also presented without any user study or quantitative evaluation of explanation quality.
minor comments (6)
  1. [Sec. 3.4, Eq. (6)] Hamming Loss is defined as 1 - Accuracy, but the standard Hamming Loss is the fraction of incorrectly predicted labels across all instances and labels. This nonstandard definition should be corrected or renamed.
  2. [Table 4] The table reports both macro and micro precision, recall, and F-measure but does not define the microaveraging formulas; please add definitions or cite a standard reference.
  3. [Listing 5] The prompt template contains typos such as "emp hasi zed_ term s" and an inconsistent feature name; these should be corrected.
  4. [Sec. 3.1] The statement that intentional concealment is reduced because the tests are embedded in the dialogues is an empirical claim with no supporting evidence; it should be softened or justified.
  5. [Sec. 4.6] The explainability module is described but not evaluated; a brief qualitative assessment or sample explanations would help verify that the generated text is coherent and faithful to the features.
  6. [Keywords] The keyword list includes "eXplainable Artificial Intelligence" and "Large Language Models" but the paper uses lowercase abbreviations; please standardize the notation throughout.

Circularity Check

1 steps flagged · score 6.0 of 10

The 90% accuracy is partially circular: the GADS/YGDS questionnaires that define the labels are embedded in the session text fed to the LLM feature extractor.

  1. self definitional [Section 3.1; Section 4.3; Listing 1]
    "Moreover, every 3 months, the chatbot uses the standard questionnaires presented in the Spanish versions of the Goldberg Anxiety and Depression Scales (GADS) and the Yesavage Geriatric Depression Scale (YGDS) to assess the cognitive state of the user. These questions are embedded during the conversation flow. The latter data is used as the label of the user (i.e., absence or presence of anxiety and depression) for the supervised learning stage. [...] Each request contains the text of the complete session and the temperature parameter set to 0."

    The GADS/YGDS answers that define the target labels are part of the session text sent to the LLM feature extractor, and Listing 1 inserts the full session at <session>. The prompt also asks the LLM to score 'anguish', 'health issues', 'negative emotion', 'sadness' and related constructs, which are the same symptom content the questionnaires measure. Consequently the engineered features can carry the label source directly, and a classifier can reach high accuracy by detecting the embedded questionnaire responses instead of learning general linguistic markers of anxiety and depression. The claimed '90% accuracy' for dialogue-based multi-label detection therefore reduces, at least in part, to giving the model the label-generating instrument as input.

full rationale

The derivation chain is mostly standard supervised ML: LLM features, feature selection, classifiers, and metrics. There is no self-citation, no imported uniqueness theorem, and no fitted parameter renamed as a prediction. However, the central claim that the system detects anxiety and depression from free dialogue is compromised by construction. Section 3.1 states that the GADS and YGDS questionnaires are embedded in the conversation flow and that their answers are used as the label; Section 4.3 and Listing 1 show that the feature extractor receives the complete session. Thus the label-generating instrument is part of the feature-generation input. The LLM is asked to score symptom constructs ('anguish', 'health issues', 'sadness', etc.) from text that contains the very questionnaire items used to define the labels. A classifier can exploit this overlap, so the reported 90% accuracy does not establish inference from natural dialogue alone. This is not a self-citation issue; it is a structural input-label overlap, which I treat as partial circularity because the prediction reduces by construction. Because the overlap affects the headline result but the rest of the pipeline is standard, the score is 6 rather than 8 or 10.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The system relies on a small set of hand-chosen hyperparameters and a few domain assumptions about label validity, feature extraction reliability, and session independence. No new physical or conceptual entities are introduced beyond the 14 named psychological and linguistic features.

free parameters (5)
  • var_smoothing (Naive Bayes) = 1e-05
    Selected via GridSearchCV over [1e-9, 1e-5, 1e-1] on the 10% cold-start subset.
  • DT hyperparameters = splitter=best, max_features=None, max_depth=100, min_samples_split=0.001, min_samples_leaf=0.001, criterion=entropy…
    Tuned via GridSearchCV on 10% of data; different values in scenario 2.
  • RF hyperparameters = n_estimators=200, max_features=sqrt, max_depth=10, min_samples_split=2, min_samples_leaf=1, criterion=gini (scenario 1)
    Tuned via GridSearchCV on 10% of data.
  • Sliding window size = 30 sessions
    Chosen by hand, not justified; used to compute average and quartiles.
  • Cold-start fraction = 10%
    Chosen by hand; used for feature selection and hyperparameter tuning.
assumptions (5)
  • domain assumption GADS and YGDS questionnaires administered by the chatbot are valid and reliable measures of anxiety and depression.
    Used as ground truth labels in Section 3.1; no external validation.
  • ad hoc to paper gpt-4o-mini accurately extracts the 14 psychological and linguistic features from dialogues.
    Feature engineering in Section 3.2 relies entirely on this LLM output without human validation.
  • domain assumption User sessions are independent observations despite coming from 32 users.
    The evaluation in Section 4.5 treats each session as an independent sample.
  • domain assumption Labels remain stable across the 3-month periods between questionnaire administrations.
    Re-labeling every 3 months (Section 3.1) assumes no within-period change in condition.
  • domain assumption Feature selection via MDI on the 10% cold-start subset generalizes to the full dataset.
    Section 4.4 uses this subset to select features for the whole system.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Detecting anxiety and depression in dialogues: a multi-label and explainable approach." pith.science (2026). https://pith.science/paper/OIPLDKCR

@misc{pith2026241217651,
  author       = {Pith},
  title        = {Pith review of: Detecting anxiety and depression in dialogues: a multi-label and explainable approach},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OIPLDKCR}},
  note         = {Machine review of arXiv:2412.17651}
}
read the original abstract

Anxiety and depression are the most common mental health issues worldwide, affecting a non-negligible part of the population. Accordingly, stakeholders, including governments' health systems, are developing new strategies to promote early detection and prevention from a holistic perspective (i.e., addressing several disorders simultaneously). In this work, an entirely novel system for the multi-label classification of anxiety and depression is proposed. The input data consists of dialogues from user interactions with an assistant chatbot. Another relevant contribution lies in using Large Language Models (LLMs) for feature extraction, provided the complexity and variability of language. The combination of LLMs, given their high capability for language understanding, and Machine Learning (ML) models, provided their contextual knowledge about the classification problem thanks to the labeled data, constitute a promising approach towards mental health assessment. To promote the solution's trustworthiness, reliability, and accountability, explainability descriptions of the model's decision are provided in a graphical dashboard. Experimental results on a real dataset attain 90 % accuracy, improving those in the prior literature. The ultimate objective is to contribute in an accessible and scalable way before formal treatment occurs in the healthcare systems.

Figures

Figures reproduced from arXiv: 2412.17651 by the authors.

Figure 1
Figure 1. System scheme. cognitive state of the end user (i.e., their emotional well-being or health condition) or on the dialogue itself (i.e., the discoursive and linguistic characteristics of the conversation with the chatbot). These features are calculated using an llm and prompt engineering, and their values range from 0.0 to 1.09 . Each generated feature is expanded with four new statistical features (average, and the t… view at source ↗
Figure 2
Figure 2. Explainable dashboard. i n s e c u r i t y : X , l o n e l i n e s s : X , n e g a t i v e _ e m o t i o n : X , p o s i t i v e _ e m o t i o n : X , s a d n e s s : X , anguish : X , h e a l t h _ i s s u e s : X , c a t a s t r o p h i c _ t e r m s : X , emphasized_terms : X , r e p e a t e d _ c o n c e p t s : X , i n t e r j e c t i o n s : X , n e g a t i v e _ a d v e r b s : X , n e g a t i v e s _ t e r m… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 19 canonical work pages

  1. [2]

    T. R. Lee, G. H. Kim, M. T. Choi, Identification of Geriatric Depression and Anxiety Using Activity Tracking Data and Minimal Geriatric Assessment Scales, Applied Sciences (Switzerland) 12 (2022)

    work page 2022

  2. [3]

    Saylam, Ö

    B. Saylam, Ö. D. İncel, Multitask Learning for Mental Health: Depression, Anxiety, Stress (DAS) Using Wearables, Diagnostics 14 (2024) 501. doi:10.3390/diagnostics14050501

    work page doi:10.3390/diagnostics14050501 2024

  3. [4]

    Marwaha, E

    S. Marwaha, E. Palmer, T. Suppes, E. Cons, A. H. Young, R. Upthegrove, Novel and emerging treatments for major depression, The Lancet 401 (2023) 141–153. doi: 10.1016/S0140-6736(22 )02080-3

    doi:10.1016/s0140-6736(22 2023

  4. [5]

    F. B. Tumaliuan, L. Grepo, E. R. Jalao, Development of Depression Data Sets and a Language Model for Depression Detection: Mixed Methods Study, JMIR Data 5 (2024) e53365. doi:10.2196/53365

    work page doi:10.2196/53365 2024

  5. [6]

    X. Wang, K. Liu, C. Wang, Knowledge-enhanced Pre-Training large language model for depression diagnosis and treatment, in: Proceeding of IEEE International Conference on Cloud Computing and Intelligence Systems, IEEE, 2023, pp. 532–536. doi:10.1109/CCIS59572.2023.10263217

    work page arXiv 2023

  6. [7]

    A. K. Chowdhury, S. R. Sujon, M. S. S. Shafi, T. Ahmmad, S. Ahmed, K. M. Hasib, F. M. Shah, Harnessing large language models over transformer models for detecting Bengali depressive social media text: A comprehensive study, Natural Language Processing Journal 7 (2024) 100075. doi:10.1016/j.nlp.2024.100075

    arXiv 2024

  7. [8]

    Nowacki, W

    A. Nowacki, W. Sitek, H. Rybiński, LLMental: Classification of Mental Disorders with Large Language Models, in: Proceedings of the International Symposium on Methodologies for Intelligent Systems, Springer, 2024, pp. 35–44. doi:10.1007/978-3-031-62700-2_4

    work page doi:10.1007/978-3-031-62700-2_4 2024

  8. [9]

    S. Ji, T. Zhang, L. Ansari, J. Fu, P. Tiwari, E. Cambria, MentalBERT: Publicly Available Pretrained Language Models for Mental Healthcare, in: Proceedings of the Language Resources and Evaluation Conference, European Language Resources Association, 2022, p. 7184–7190

    work page 2022

Show all 35 references
  1. [10]

    K. Yang, T. Zhang, Z. Kuang, Q. Xie, J. Huang, S. Ananiadou, MentaLLaMA: interpretable mental health analysis on social media with large language models, in: Proceedings of the ACM on Web Conference, Association for Computing Machinery, 2024, pp. 4489–4500. doi:10.1145/3589334...

    2024 doi

  2. [11]

    Vajre, M

    V. Vajre, M. Naylor, U. Kamath, A. Shehu, PsychBERT: A Mental Health Language Model for Social Media Mental Health Behavioral Analysis, in: Proceedings of the IEEE International Conference on Bioinformatics and Biomedicine, IEEE, 2021, pp. 1077–1082. doi:10.1109/BIBM52615.2021...

    2021

  3. [12]

    T. Wang, A. Roberts, D. Hesslow, T. L. Scao, H. W. Chung, I. Beltagy, J. Launay, C. Raffel, What Language Model Architecture and Pretraining Objective Work Best for Zero-Shot Generalization?, in: Proceedings of Machine Learning Research, volume 162, MLR Press, 2022, pp. 1–21

    2022

  4. [13]

    M. Moor, O. Banerjee, Z. S. H. Abad, H. M. Krumholz, J. Leskovec, E. J. Topol, P. Rajpurkar, Foundation models for generalist medical artificial intelligence, Nature 616 (2023) 259–265. doi:10.1038/s41586-023-05881-4

    2023 doi

  5. [14]

    Y. Liu, X. Ding, S. Peng, C. Zhang, Leveraging ChatGPT to optimize depression intervention through explainable deep learning, Frontiers in Psychiatry 15 (2024) 1383648. doi: 10.3389/fpsy t.2024.1383648

    2024

  6. [15]

    X. Xu, B. Yao, Y. Dong, S. Gabriel, H. Yu, J. Hendler, M. Ghassemi, A. K. Dey, D. Wang, Mental-llm: Leveraging large language models for mental health prediction via online text data, in: Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, volu...

    2024 doi

  7. [16]

    Gomes, M

    N. Gomes, M. Pato, A. R. Lourenço, N. Datia, A Survey on Wearable Sensors for Mental Health Monitoring, Sensors 23 (2023) 1330. doi: 10.3390/s23031330

    2023 doi

  8. [17]

    A. M. Salih, Z. Raisi-Estabragh, I. B. Galazzo, P. Radeva, S. E. Petersen, K. Lekadir, G. Menegaz, A Perspective on Explainable Artificial Intelligence Methods: SHAP and LIME, Advanced Intelligent Systems (2024) 2400304. doi:10.1002/aisy.202400304

    2024 doi

  9. [18]

    Coroama, A

    L. Coroama, A. Groza, Explainable Artificial Intelligence for Person Identification, in: Proceedings of the IEEE International Conference on Intelligent Computer Communication and Processing, IEEE, 2021, pp. 375–382. doi:10.1109/ICCP53602.2021.9733525

    2021

  10. [19]

    Nannini, J

    L. Nannini, J. Alonso-Moral, A. Catala, M. Lama, S. Barro, Operationalizing Explainable AI in the EU Regulatory Ecosystem, IEEE Intelligent Systems (2024) 37–48. doi:10.1109/MIS.2024.3383155

    2024

  11. [20]

    L. Ren, H. Lin, B. Xu, S. Zhang, L. Yang, S. Sun, Depression Detection on Reddit With an Emotion- Based Attention Network: Algorithm Development and Validation, JMIR Medical Informatics 9 (2021) e28754. doi:10.2196/28754

    2021 doi

  12. [21]

    A. B. S. Rahman, H.-T. Ta, L. Najjar, A. Azadmanesh, A. S. Gönul, DepressionEmo: A novel dataset for multilabel classification of depression emotions, Journal of Affective Disorders (2024) 445–458. doi:10.1016/j.jad.2024.08.013

    2024 doi

  13. [22]

    R. W. Levenson, Stress and Illness: A Role for Specific Emotions, Psychosomatic Medicine 81 (2019) 720–730. doi:10.1097/PSY.0000000000000736

    2019 doi

  14. [23]

    S. A. Qureshi, G. Dias, M. Hasanuzzaman, S. Saha, Improving Depression Level Estimation by Concurrently Learning Emotion Intensity, IEEE Computational Intelligence Magazine 15 (2020) 47–59. doi:10.1109/MCI.2020.2998234

    2020

  15. [24]

    Ghosh, A

    S. Ghosh, A. Ekbal, P. Bhattacharyya, What Does Your Bio Say? Inferring Twitter Users’ Depres- sion Status From Multimodal Profile Information Using Deep Learning, IEEE Transactions on Computational Social Systems 9 (2022) 1484–1494. doi:10.1109/TCSS.2021.3116242

    2022

  16. [25]

    Turcan, S

    E. Turcan, S. Muresan, K. McKeown, Emotion-Infused Models for Explainable Psychological Stress Detection, in: Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Association for Computationa...

    2021 doi

  17. [26]

    Ghosh, A

    S. Ghosh, A. Ekbal, P. Bhattacharyya, A Multitask Framework to Detect Depression, Sentiment and Multi-label Emotion from Suicide Notes, Cognitive Computation 14 (2022) 110–129. doi: 10.1 007/s12559-021-09828-7

    2022

  18. [27]

    Sarkar, A

    S. Sarkar, A. Alhamadani, L. Alkulaib, C.-T. Lu, Predicting Depression and Anxiety on Reddit: a Multi-task Learning Approach, in: Proceedings of the IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining, IEEE, 2022, pp. 427–435. doi:10.1109/ASON ...

    2022

  19. [28]

    D. Park, G. Lee, S. Kim, T. Seo, H. Oh, S. J. Kim, Probability-based multi-label classification considering correlation between labels–focusing on DSM-5 depressive disorder diagnostic criteria, IEEE Access (2024) 70289–70296. doi:10.1109/ACCESS.2024.3401704

    2024

  20. [29]

    Ilias, D

    L. Ilias, D. Askounis, Multitask learning for recognizing stress and depression in social media, Online Social Networks and Media 37-38 (2023) 100270. doi:10.1016/j.osnem.2023.100270

    2023

  21. [30]

    D. Park, S. Lim, Y. Choi, H. Oh, Depression Emotion Multi-Label Classification Using Everytime Platform With DSM-5 Diagnostic Criteria, IEEE Access 11 (2023) 89093–89106. doi: 10.1109/AC CESS.2023.3305477

    2023

  22. [31]

    V. B. de Souza, J. C. Nobre, K. Becker, DAC Stacking: A Deep Learning Ensemble to Classify Anxiety, Depression, and Their Comorbidity From Reddit Texts, IEEE Journal of Biomedical and Health Informatics 26 (2022) 3303–3311. doi:10.1109/JBHI.2022.3151589

    2022

  23. [32]

    Ilias, S

    L. Ilias, S. Mouzakitis, D. Askounis, Calibration of Transformer-Based Models for Identifying Stress and Depression in Social Media, IEEE Transactions on Computational Social Systems 11 (2024) 1979–1990. doi:10.1109/TCSS.2023.3283009

    2024

  24. [33]

    J. Ohse, B. Hadžić, P. Mohammed, N. Peperkorn, M. Danner, A. Yorita, N. Kubota, M. Rätsch, Y. Shiban, Zero-Shot Strike: Testing the generalisation capabilities of out-of-the-box LLM models for depression detection, Computer Speech & Language 88 (2024) 101663. doi: 10.1016/j.cs...

    2024

  25. [34]

    Y. Wang, D. Inkpen, P. K. Gamaarachchige, Explainable depression detection using large language models on social media data, in: Proceedings of the Workshop on Computational Linguistics and Clinical Psychology, Association for Computational Linguistics, 2024, pp. 108–126

    2024

  26. [35]

    Rivolli, J

    A. Rivolli, J. Read, C. Soares, B. Pfahringer, A. C. P. L. F. de Carvalho, An empirical analysis of binary transformation strategies and base algorithms for multi-label learning, Machine Learning 109 (2020) 1509–1563. doi:10.1007/s10994-020-05879-3

    2020 doi

  27. [2488]

    doi:10.3390/app12052488

    doi

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.