Towards Explainable Depression Detection: A Neurosymbolic Approach to Uncover Social Media Signals with Generative AI

Mohammad Saeid Mahdavinejad, Peyman Adibi, Amirhassan Monajemi, Pascal Hitzler
Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning, PMLR 284:830-853, 2025.

Abstract

Depression remains a pervasive mental health disorder that demands prompt diagnosis and intervention. Although social media data presents a promising avenue for early detection, traditional deep neural models are frequently critiqued for their lack of interpretability and susceptibility to bias. We introduce ProtoDep—a neurosymbolic framework that integrates clinically grounded categorizations (e.g., PHQ-9 symptoms) with large language model–assisted prototype learning. Unlike conventional black-box models, ProtoDep aligns individual tweets with symptom-level prototypes, offering interpretable explanations at three levels: (i) symptom-level insights that map user posts to recognized depressive patterns, (ii) case-based reasoning that compares users to representative prototype profiles, and (iii) transparent concept-level decisions, wherein classification at inference time is driven by the distances between the user profile and prototype user and symptom clusters, yielding clear, quantifiable explanations. By integrating symbolic mental health constructs with neural embeddings, ProtoDep achieves a mean F1-score of 94% across five benchmark datasets and establishes a foundation for interpretable depression screening pipelines with potential applicability in clinical settings.

Cite this Paper

BibTeX
@InProceedings{pmlr-v284-mahdavinejad25a, title = {Towards Explainable Depression Detection: A Neurosymbolic Approach to Uncover Social Media Signals with Generative AI}, author = {Mahdavinejad, Mohammad Saeid and Adibi, Peyman and Monajemi, Amirhassan and Hitzler, Pascal}, booktitle = {Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning}, pages = {830--853}, year = {2025}, editor = {H. Gilpin, Leilani and Giunchiglia, Eleonora and Hitzler, Pascal and van Krieken, Emile}, volume = {284}, series = {Proceedings of Machine Learning Research}, month = {08--10 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v284/main/assets/mahdavinejad25a/mahdavinejad25a.pdf}, url = {https://proceedings.mlr.press/v284/mahdavinejad25a.html}, abstract = {Depression remains a pervasive mental health disorder that demands prompt diagnosis and intervention. Although social media data presents a promising avenue for early detection, traditional deep neural models are frequently critiqued for their lack of interpretability and susceptibility to bias. We introduce ProtoDep—a neurosymbolic framework that integrates clinically grounded categorizations (e.g., PHQ-9 symptoms) with large language model–assisted prototype learning. Unlike conventional black-box models, ProtoDep aligns individual tweets with symptom-level prototypes, offering interpretable explanations at three levels: (i) symptom-level insights that map user posts to recognized depressive patterns, (ii) case-based reasoning that compares users to representative prototype profiles, and (iii) transparent concept-level decisions, wherein classification at inference time is driven by the distances between the user profile and prototype user and symptom clusters, yielding clear, quantifiable explanations. By integrating symbolic mental health constructs with neural embeddings, ProtoDep achieves a mean F1-score of 94% across five benchmark datasets and establishes a foundation for interpretable depression screening pipelines with potential applicability in clinical settings.} }
Endnote
%0 Conference Paper %T Towards Explainable Depression Detection: A Neurosymbolic Approach to Uncover Social Media Signals with Generative AI %A Mohammad Saeid Mahdavinejad %A Peyman Adibi %A Amirhassan Monajemi %A Pascal Hitzler %B Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning %C Proceedings of Machine Learning Research %D 2025 %E Leilani H. Gilpin %E Eleonora Giunchiglia %E Pascal Hitzler %E Emile van Krieken %F pmlr-v284-mahdavinejad25a %I PMLR %P 830--853 %U https://proceedings.mlr.press/v284/mahdavinejad25a.html %V 284 %X Depression remains a pervasive mental health disorder that demands prompt diagnosis and intervention. Although social media data presents a promising avenue for early detection, traditional deep neural models are frequently critiqued for their lack of interpretability and susceptibility to bias. We introduce ProtoDep—a neurosymbolic framework that integrates clinically grounded categorizations (e.g., PHQ-9 symptoms) with large language model–assisted prototype learning. Unlike conventional black-box models, ProtoDep aligns individual tweets with symptom-level prototypes, offering interpretable explanations at three levels: (i) symptom-level insights that map user posts to recognized depressive patterns, (ii) case-based reasoning that compares users to representative prototype profiles, and (iii) transparent concept-level decisions, wherein classification at inference time is driven by the distances between the user profile and prototype user and symptom clusters, yielding clear, quantifiable explanations. By integrating symbolic mental health constructs with neural embeddings, ProtoDep achieves a mean F1-score of 94% across five benchmark datasets and establishes a foundation for interpretable depression screening pipelines with potential applicability in clinical settings.
APA
Mahdavinejad, M.S., Adibi, P., Monajemi, A. & Hitzler, P.. (2025). Towards Explainable Depression Detection: A Neurosymbolic Approach to Uncover Social Media Signals with Generative AI. Proceedings of The 19th International Conference on Neurosymbolic Learning and Reasoning, in Proceedings of Machine Learning Research 284:830-853 Available from https://proceedings.mlr.press/v284/mahdavinejad25a.html.

Related Material