Supervised Learning and Large Language Model Benchmarks on Mental Health Datasets: Cognitive Distortions and Suicidal Risks in Chinese Social Media

On social media, users often express their personal feelings, which may exhibit cognitive distortions or even suicidal tendencies on certain specific topics. Early recognition of these signs is critical for effective psychological intervention. In this paper, we introduce two novel datasets from Chi...

Full description

Saved in:
Bibliographic Details
Published inBioengineering (Basel) Vol. 12; no. 8; p. 882
Main Authors Qi, Hongzhi, Fu, Guanghui, Li, Jianqiang, Song, Changwei, Zhai, Wei, Luo, Dan, Liu, Shuo, Yu, Yijing, Yang, Bingxiang, Zhao, Qing
Format Journal Article
LanguageEnglish
Published Switzerland MDPI AG 19.08.2025
MDPI
Subjects
Online AccessGet full text
ISSN2306-5354
2306-5354
DOI10.3390/bioengineering12080882

Cover

Loading…
Abstract On social media, users often express their personal feelings, which may exhibit cognitive distortions or even suicidal tendencies on certain specific topics. Early recognition of these signs is critical for effective psychological intervention. In this paper, we introduce two novel datasets from Chinese social media: SOS-HL-1K for suicidal risk classification, which contains 1249 posts, and SocialCD-3K, a multi-label classification dataset for cognitive distortion detection that contains 3407 posts. We conduct a comprehensive evaluation using two supervised learning methods and eight large language models (LLMs) on the proposed datasets. From the prompt engineering perspective, we experiment with two types of prompt strategies, including four zero-shot and five few-shot strategies. We also evaluate the performance of the LLMs after fine-tuning on the proposed tasks. Experimental results show a significant performance gap between prompted LLMs and supervised learning. Our best supervised model achieves strong results, with an F1-score of 82.76% for the high-risk class in the suicide task and a micro-averaged F1-score of 76.10% for the cognitive distortion task. Without fine-tuning, the best-performing LLM lags by 6.95 percentage points in the suicide task and a more pronounced 31.53 points in the cognitive distortion task. Fine-tuning substantially narrows this performance gap to 4.31% and 3.14% for the respective tasks. While this research highlights the potential of LLMs in psychological contexts, it also shows that supervised learning remains necessary for more challenging tasks.
AbstractList On social media, users often express their personal feelings, which may exhibit cognitive distortions or even suicidal tendencies on certain specific topics. Early recognition of these signs is critical for effective psychological intervention. In this paper, we introduce two novel datasets from Chinese social media: SOS-HL-1K for suicidal risk classification, which contains 1249 posts, and SocialCD-3K, a multi-label classification dataset for cognitive distortion detection that contains 3407 posts. We conduct a comprehensive evaluation using two supervised learning methods and eight large language models (LLMs) on the proposed datasets. From the prompt engineering perspective, we experiment with two types of prompt strategies, including four zero-shot and five few-shot strategies. We also evaluate the performance of the LLMs after fine-tuning on the proposed tasks. Experimental results show a significant performance gap between prompted LLMs and supervised learning. Our best supervised model achieves strong results, with an F1-score of 82.76% for the high-risk class in the suicide task and a micro-averaged F1-score of 76.10% for the cognitive distortion task. Without fine-tuning, the best-performing LLM lags by 6.95 percentage points in the suicide task and a more pronounced 31.53 points in the cognitive distortion task. Fine-tuning substantially narrows this performance gap to 4.31% and 3.14% for the respective tasks. While this research highlights the potential of LLMs in psychological contexts, it also shows that supervised learning remains necessary for more challenging tasks.
On social media, users often express their personal feelings, which may exhibit cognitive distortions or even suicidal tendencies on certain specific topics. Early recognition of these signs is critical for effective psychological intervention. In this paper, we introduce two novel datasets from Chinese social media: SOS-HL-1K for suicidal risk classification, which contains 1249 posts, and SocialCD-3K, a multi-label classification dataset for cognitive distortion detection that contains 3407 posts. We conduct a comprehensive evaluation using two supervised learning methods and eight large language models (LLMs) on the proposed datasets. From the prompt engineering perspective, we experiment with two types of prompt strategies, including four zero-shot and five few-shot strategies. We also evaluate the performance of the LLMs after fine-tuning on the proposed tasks. Experimental results show a significant performance gap between prompted LLMs and supervised learning. Our best supervised model achieves strong results, with an F1-score of 82.76% for the high-risk class in the suicide task and a micro-averaged F1-score of 76.10% for the cognitive distortion task. Without fine-tuning, the best-performing LLM lags by 6.95 percentage points in the suicide task and a more pronounced 31.53 points in the cognitive distortion task. Fine-tuning substantially narrows this performance gap to 4.31% and 3.14% for the respective tasks. While this research highlights the potential of LLMs in psychological contexts, it also shows that supervised learning remains necessary for more challenging tasks.On social media, users often express their personal feelings, which may exhibit cognitive distortions or even suicidal tendencies on certain specific topics. Early recognition of these signs is critical for effective psychological intervention. In this paper, we introduce two novel datasets from Chinese social media: SOS-HL-1K for suicidal risk classification, which contains 1249 posts, and SocialCD-3K, a multi-label classification dataset for cognitive distortion detection that contains 3407 posts. We conduct a comprehensive evaluation using two supervised learning methods and eight large language models (LLMs) on the proposed datasets. From the prompt engineering perspective, we experiment with two types of prompt strategies, including four zero-shot and five few-shot strategies. We also evaluate the performance of the LLMs after fine-tuning on the proposed tasks. Experimental results show a significant performance gap between prompted LLMs and supervised learning. Our best supervised model achieves strong results, with an F1-score of 82.76% for the high-risk class in the suicide task and a micro-averaged F1-score of 76.10% for the cognitive distortion task. Without fine-tuning, the best-performing LLM lags by 6.95 percentage points in the suicide task and a more pronounced 31.53 points in the cognitive distortion task. Fine-tuning substantially narrows this performance gap to 4.31% and 3.14% for the respective tasks. While this research highlights the potential of LLMs in psychological contexts, it also shows that supervised learning remains necessary for more challenging tasks.
Audience Academic
Author Luo, Dan
Qi, Hongzhi
Zhao, Qing
Fu, Guanghui
Song, Changwei
Yang, Bingxiang
Yu, Yijing
Li, Jianqiang
Zhai, Wei
Liu, Shuo
AuthorAffiliation 3 School of Nursing, Wuhan University, Wuhan 430071, China
2 Institut du Cerveau—Paris Brain Institute-ICM, Sorbonne Université, CNRS, Inria, Inserm, AP-HP, Hôpital de la Pitié-Salpêtrière, 75013 Paris, France
1 College of Computer Science, Beijing University of Technology, Beijing 100124, China; qhz123@emails.bjut.edu.cn (H.Q.)
AuthorAffiliation_xml – name: 2 Institut du Cerveau—Paris Brain Institute-ICM, Sorbonne Université, CNRS, Inria, Inserm, AP-HP, Hôpital de la Pitié-Salpêtrière, 75013 Paris, France
– name: 3 School of Nursing, Wuhan University, Wuhan 430071, China
– name: 1 College of Computer Science, Beijing University of Technology, Beijing 100124, China; qhz123@emails.bjut.edu.cn (H.Q.)
Author_xml – sequence: 1
  givenname: Hongzhi
  orcidid: 0009-0004-8007-3257
  surname: Qi
  fullname: Qi, Hongzhi
– sequence: 2
  givenname: Guanghui
  surname: Fu
  fullname: Fu, Guanghui
– sequence: 3
  givenname: Jianqiang
  orcidid: 0000-0003-1995-9249
  surname: Li
  fullname: Li, Jianqiang
– sequence: 4
  givenname: Changwei
  surname: Song
  fullname: Song, Changwei
– sequence: 5
  givenname: Wei
  surname: Zhai
  fullname: Zhai, Wei
– sequence: 6
  givenname: Dan
  surname: Luo
  fullname: Luo, Dan
– sequence: 7
  givenname: Shuo
  surname: Liu
  fullname: Liu, Shuo
– sequence: 8
  givenname: Yijing
  surname: Yu
  fullname: Yu, Yijing
– sequence: 9
  givenname: Bingxiang
  surname: Yang
  fullname: Yang, Bingxiang
– sequence: 10
  givenname: Qing
  surname: Zhao
  fullname: Zhao, Qing
BackLink https://www.ncbi.nlm.nih.gov/pubmed/40868395$$D View this record in MEDLINE/PubMed
BookMark eNptkt9u0zAUxiM0xMbYK0yWuOGmw_-SONyg0QGb1AqJwrXl2CepS2oPO6nEc_DCnK5j2hCKlNgnn38-39H3sjgKMUBRnDN6IURD37Y-Quh9AEg-9IxTRZXiz4oTLmg1K0Upjx6tj4uznDeUUiZ4ySv5ojiWVFVKNOVJ8Xs13ULa-QyOLMCkgEBiAm5M6gHfoZ8MLpbRwUA-QLDrrUk_MomBLCGMZiDXYIZxTa7MaDKM-R2Zxz740e-AXPk8xjT6GPIddDV56x2e-eozMnwg8zXayEBW0XqsL8F586p43pkhw9n997T4_unjt_n1bPHl8838cjGzopZ8JoFT2wmpmOga2qrWslqiSYC2Y6ytpaDONbWVtJUUOt5KcNR0rt6PgrJanBY3B66LZqNvk0dnv3Q0Xt8VYuq1webtAFp2YE1Ja8NcLaVyrTDU4F2yYoorBsh6f2DdTu0WnMXRJDM8gT79E_xa93GnGRdKKFoh4c09IcWfE-RRb322MAwmQJyyFlxWkgla7xt__Y90E6cUcFZ7lWiwpXIPvDioeoMOfOgiXmzxcbD1FhPVeaxfKsxIVTWswQPnjz08NP83LSioDgKbYs4JugcJo3qfTP3_ZIo_xj3ajQ
Cites_doi 10.1613/jair.1.13566
10.18653/v1/2025.clpsych-1.17
10.1007/s00330-023-10213-1
10.1109/ICDEW67478.2025.00029
10.2196/57234
10.1007/s13278-021-00737-z
10.1109/JIOT.2020.3004376
10.1177/1178222618792860
10.1111/eip.12229
10.2196/26119
10.1016/j.lindif.2023.102274
10.1007/s13278-021-00776-6
10.1109/ICHI.2017.47
10.1038/s41586-023-06160-y
10.2196/preprints.52656
10.1002/eng2.12189
10.18653/v1/D19-1044
10.1109/CISCE.2019.00146
10.1038/s42256-023-00626-4
10.3350/cmh.2023.0089
10.2105/AJPH.2011.300608
10.1016/S2589-7500(23)00019-5
10.18653/v1/2023.findings-emnlp.680
10.1038/s41562-024-01847-2
10.1080/02673843.2019.1590851
10.1016/S2215-0366(18)30511-X
10.18653/v1/2023.emnlp-main.370
10.1093/jamiaopen/ooaa010
10.18653/v1/2023.acl-long.555
10.1038/s41591-023-02448-8
ContentType Journal Article
Copyright COPYRIGHT 2025 MDPI AG
2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
2025 by the authors. 2025
Copyright_xml – notice: COPYRIGHT 2025 MDPI AG
– notice: 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
– notice: 2025 by the authors. 2025
DBID AAYXX
CITATION
NPM
8FE
8FG
8FH
ABJCF
ABUWG
AFKRA
AZQEC
BBNVY
BENPR
BGLVJ
BHPHI
CCPQU
DWQXO
GNUQQ
HCIFZ
L6V
LK8
M7P
M7S
PHGZM
PHGZT
PIMPY
PKEHL
PQEST
PQGLB
PQQKQ
PQUKI
PTHSS
7X8
5PM
DOA
DOI 10.3390/bioengineering12080882
DatabaseName CrossRef
PubMed
ProQuest SciTech Collection
ProQuest Technology Collection
ProQuest Natural Science Collection
Materials Science & Engineering Collection
ProQuest Central (Alumni)
ProQuest Central UK/Ireland
ProQuest Central Essentials
Biological Science Collection
ProQuest Central
Technology Collection
Natural Science Collection
ProQuest One Community College
ProQuest Central Korea
ProQuest Central Student
SciTech Premium Collection
ProQuest Engineering Collection
Biological Sciences
Biological Science Database
Engineering Database
ProQuest Central Premium
ProQuest One Academic (New)
Publicly Available Content Database
ProQuest One Academic Middle East (New)
ProQuest One Academic Eastern Edition (DO NOT USE)
ProQuest One Applied & Life Sciences
ProQuest One Academic
ProQuest One Academic UKI Edition
Engineering Collection
MEDLINE - Academic
PubMed Central (Full Participant titles)
DOAJ - Directory of Open Access Journals (Some content may be blocked by TCTC IT security protocols)
DatabaseTitle CrossRef
PubMed
Publicly Available Content Database
ProQuest Central Student
Technology Collection
ProQuest One Academic Middle East (New)
ProQuest Central Essentials
ProQuest Central (Alumni Edition)
SciTech Premium Collection
ProQuest One Community College
ProQuest Natural Science Collection
ProQuest Central
ProQuest One Applied & Life Sciences
ProQuest Engineering Collection
Natural Science Collection
ProQuest Central Korea
Biological Science Collection
ProQuest Central (New)
Engineering Collection
Engineering Database
ProQuest Biological Science Collection
ProQuest One Academic Eastern Edition
ProQuest Technology Collection
Biological Science Database
ProQuest SciTech Collection
ProQuest One Academic UKI Edition
Materials Science & Engineering Collection
ProQuest One Academic
ProQuest One Academic (New)
MEDLINE - Academic
DatabaseTitleList Publicly Available Content Database
PubMed


MEDLINE - Academic

CrossRef
Database_xml – sequence: 1
  dbid: DOA
  name: DOAJ Directory of Open Access Journals
  url: https://www.doaj.org/
  sourceTypes: Open Website
– sequence: 2
  dbid: NPM
  name: PubMed
  url: https://proxy.k.utb.cz/login?url=http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 3
  dbid: 8FG
  name: ProQuest Technology Collection
  url: https://search.proquest.com/technologycollection1
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
EISSN 2306-5354
ExternalDocumentID oai_doaj_org_article_4feca507a1d7448db3a0a0134618281e
PMC12383806
A853566919
40868395
10_3390_bioengineering12080882
Genre Journal Article
GeographicLocations United Kingdom
China
GeographicLocations_xml – name: United Kingdom
– name: China
GrantInformation_xml – fundername: National Natural Science Foundation of China
  grantid: 72174152 and 72474166
– fundername: Beijing Natural Science Foundation
  grantid: 7254302
– fundername: Fundamental Research Funds for the Central Universities
  grantid: 2042022kf1218 and 2042022kf1037
– fundername: Young Top-notch Talent Cultivation Program of Hubei Province
– fundername: Fundamental Research Funds for the Central Universities
  grantid: 2042022kf1218; 2042022kf1037
– fundername: National Natural Science Foundation of China
  grantid: 72174152; 72474166
GroupedDBID 53G
5VS
8FE
8FG
8FH
AAFWJ
AAYXX
ABDBF
ABJCF
ACUHS
ADBBV
AFKRA
AFPKN
ALMA_UNASSIGNED_HOLDINGS
AOIJS
BBNVY
BCNDV
BENPR
BGLVJ
BHPHI
CCPQU
CITATION
GROUPED_DOAJ
HCIFZ
HYE
IAO
IHR
INH
ITC
KQ8
L6V
LK8
M7P
M7S
MODMG
M~E
OK1
PGMZT
PHGZM
PHGZT
PIMPY
PQGLB
PROAC
PTHSS
PUEGO
RPM
NPM
ABUWG
AZQEC
DWQXO
GNUQQ
PKEHL
PQEST
PQQKQ
PQUKI
7X8
5PM
ID FETCH-LOGICAL-c3742-4e20cf34813f90b8bc174013eebf11b7430dd97c40b40ef2b4ed0afd700010173
IEDL.DBID DOA
ISSN 2306-5354
IngestDate Mon Sep 01 19:38:54 EDT 2025
Fri Aug 29 05:29:48 EDT 2025
Thu Sep 04 12:40:56 EDT 2025
Wed Aug 27 13:32:43 EDT 2025
Tue Sep 02 03:58:31 EDT 2025
Thu Sep 04 05:00:39 EDT 2025
Wed Aug 27 16:39:40 EDT 2025
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 8
Keywords deep learning
large language model
suicide detection
social media
mental health
cognitive distortions
Language English
License https://creativecommons.org/licenses/by/4.0
Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c3742-4e20cf34813f90b8bc174013eebf11b7430dd97c40b40ef2b4ed0afd700010173
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ORCID 0000-0003-1995-9249
0009-0004-8007-3257
OpenAccessLink https://doaj.org/article/4feca507a1d7448db3a0a0134618281e
PMID 40868395
PQID 3243982856
PQPubID 2055440
ParticipantIDs doaj_primary_oai_doaj_org_article_4feca507a1d7448db3a0a0134618281e
pubmedcentral_primary_oai_pubmedcentral_nih_gov_12383806
proquest_miscellaneous_3246413077
proquest_journals_3243982856
gale_infotracacademiconefile_A853566919
pubmed_primary_40868395
crossref_primary_10_3390_bioengineering12080882
PublicationCentury 2000
PublicationDate 20250819
PublicationDateYYYYMMDD 2025-08-19
PublicationDate_xml – month: 8
  year: 2025
  text: 20250819
  day: 19
PublicationDecade 2020
PublicationPlace Switzerland
PublicationPlace_xml – name: Switzerland
– name: Basel
PublicationTitle Bioengineering (Basel)
PublicationTitleAlternate Bioengineering (Basel)
PublicationYear 2025
Publisher MDPI AG
MDPI
Publisher_xml – name: MDPI AG
– name: MDPI
References ref_50
Acheampong (ref_9) 2020; 2
ref_14
ref_58
ref_13
ref_11
ref_55
ref_54
ref_53
Srivastava (ref_60) 2014; 15
ref_51
Huang (ref_2) 2019; 6
Keles (ref_4) 2020; 25
ref_19
ref_18
Laparra (ref_12) 2020; 3
ref_17
ref_16
ref_59
Saunders (ref_10) 2022; 75
Vaswani (ref_46) 2017; 30
ref_25
ref_24
ref_23
ref_22
Coppersmith (ref_7) 2018; 10
ref_21
ref_20
Zhang (ref_29) 2020; 7
Lin (ref_56) 2024; 8
ref_28
Yeo (ref_36) 2023; 29
Robinson (ref_5) 2016; 10
Brown (ref_52) 2020; 33
Thirunavukarasu (ref_15) 2023; 29
Luxton (ref_6) 2012; 102
Surameery (ref_34) 2023; 3
ref_33
ref_31
ref_30
Kasneci (ref_35) 2023; 103
ref_39
Jiang (ref_37) 2023; 619
Nandwani (ref_8) 2021; 11
Liebrenz (ref_32) 2023; 5
ref_47
ref_45
ref_44
ref_43
ref_42
ref_41
ref_40
ref_1
Fu (ref_26) 2021; 23
ref_3
Bauer (ref_38) 2024; 11
ref_49
ref_48
Singh (ref_27) 2021; 11
Ding (ref_57) 2023; 5
References_xml – volume: 75
  start-page: 351
  year: 2022
  ident: ref_10
  article-title: Domain adaptation and multi-domain adaptation for neural machine translation: A survey
  publication-title: J. Artif. Intell. Res.
  doi: 10.1613/jair.1.13566
– volume: 30
  start-page: 5998
  year: 2017
  ident: ref_46
  article-title: Attention is all you need
  publication-title: Adv. Neural Inf. Process. Syst.
– ident: ref_49
– ident: ref_55
– ident: ref_22
  doi: 10.18653/v1/2025.clpsych-1.17
– ident: ref_33
  doi: 10.1007/s00330-023-10213-1
– ident: ref_51
– ident: ref_23
  doi: 10.1109/ICDEW67478.2025.00029
– volume: 11
  start-page: e57234
  year: 2024
  ident: ref_38
  article-title: Using Large Language Models to Understand Suicidality in a Social Media–Based Taxonomy of Mental Health Disorders: Linguistic Analysis of Reddit Posts
  publication-title: JMIR Ment. Health
  doi: 10.2196/57234
– volume: 11
  start-page: 33
  year: 2021
  ident: ref_27
  article-title: Sentiment analysis on the impact of coronavirus in social life using the BERT model
  publication-title: Soc. Netw. Anal. Min.
  doi: 10.1007/s13278-021-00737-z
– volume: 7
  start-page: 8133
  year: 2020
  ident: ref_29
  article-title: Emotion detection in online social networks: A multilabel learning approach
  publication-title: IEEE Internet Things J.
  doi: 10.1109/JIOT.2020.3004376
– ident: ref_16
– ident: ref_39
– volume: 10
  start-page: 1178222618792860
  year: 2018
  ident: ref_7
  article-title: Natural language processing of social media as screening for suicide risk
  publication-title: Biomed. Inform. Insights
  doi: 10.1177/1178222618792860
– ident: ref_1
– volume: 10
  start-page: 103
  year: 2016
  ident: ref_5
  article-title: Social media and suicide prevention: A systematic review
  publication-title: Early Interv. Psychiatry
  doi: 10.1111/eip.12229
– ident: ref_58
– volume: 23
  start-page: e26119
  year: 2021
  ident: ref_26
  article-title: Distant supervision for mental health management in social media: Suicide risk classification system development study
  publication-title: J. Med. Internet Res.
  doi: 10.2196/26119
– ident: ref_31
– volume: 15
  start-page: 1929
  year: 2014
  ident: ref_60
  article-title: Dropout: A simple way to prevent neural networks from overfitting
  publication-title: J. Mach. Learn. Res.
– ident: ref_48
– volume: 103
  start-page: 102274
  year: 2023
  ident: ref_35
  article-title: ChatGPT for good? On opportunities and challenges of large language models for education
  publication-title: Learn. Individ. Differ.
  doi: 10.1016/j.lindif.2023.102274
– ident: ref_41
– volume: 11
  start-page: 81
  year: 2021
  ident: ref_8
  article-title: A review on sentiment analysis and emotion detection from text
  publication-title: Soc. Netw. Anal. Min.
  doi: 10.1007/s13278-021-00776-6
– ident: ref_25
  doi: 10.1109/ICHI.2017.47
– volume: 619
  start-page: 357
  year: 2023
  ident: ref_37
  article-title: Health system-scale language models are all-purpose prediction engines
  publication-title: Nature
  doi: 10.1038/s41586-023-06160-y
– ident: ref_42
  doi: 10.2196/preprints.52656
– ident: ref_13
– volume: 33
  start-page: 1877
  year: 2020
  ident: ref_52
  article-title: Language models are few-shot learners
  publication-title: Adv. Neural Inf. Process. Syst.
– ident: ref_17
– ident: ref_45
– volume: 2
  start-page: e12189
  year: 2020
  ident: ref_9
  article-title: Text-based emotion detection: Advances, challenges, and opportunities
  publication-title: Eng. Rep.
  doi: 10.1002/eng2.12189
– ident: ref_20
– ident: ref_44
  doi: 10.18653/v1/D19-1044
– ident: ref_59
– ident: ref_28
  doi: 10.1109/CISCE.2019.00146
– ident: ref_53
– ident: ref_30
– volume: 5
  start-page: 220
  year: 2023
  ident: ref_57
  article-title: Parameter-efficient fine-tuning of large-scale pre-trained language models
  publication-title: Nat. Mach. Intell.
  doi: 10.1038/s42256-023-00626-4
– volume: 3
  start-page: 17
  year: 2023
  ident: ref_34
  article-title: Use ChatGPT to solve programming bugs
  publication-title: Int. J. Inf. Technol. Comput. Eng. (IJITC)
– ident: ref_3
– ident: ref_24
– volume: 29
  start-page: 721
  year: 2023
  ident: ref_36
  article-title: Assessing the performance of ChatGPT in answering questions regarding cirrhosis and hepatocellular carcinoma
  publication-title: Clin. Mol. Hepatol.
  doi: 10.3350/cmh.2023.0089
– ident: ref_47
– ident: ref_40
– volume: 102
  start-page: S195
  year: 2012
  ident: ref_6
  article-title: Social media and suicide: A public health perspective
  publication-title: Am. J. Public Health
  doi: 10.2105/AJPH.2011.300608
– ident: ref_14
– volume: 5
  start-page: e105
  year: 2023
  ident: ref_32
  article-title: Generating scholarly content with ChatGPT: Ethical challenges for medical publishing
  publication-title: Lancet Digit. Health
  doi: 10.1016/S2589-7500(23)00019-5
– ident: ref_21
– ident: ref_18
  doi: 10.18653/v1/2023.findings-emnlp.680
– ident: ref_50
– ident: ref_54
– volume: 8
  start-page: 611
  year: 2024
  ident: ref_56
  article-title: How to write effective prompts for large language models
  publication-title: Nat. Hum. Behav.
  doi: 10.1038/s41562-024-01847-2
– volume: 25
  start-page: 79
  year: 2020
  ident: ref_4
  article-title: A systematic review: The influence of social media on depression, anxiety and psychological distress in adolescents
  publication-title: Int. J. Adolesc. Youth
  doi: 10.1080/02673843.2019.1590851
– volume: 6
  start-page: 211
  year: 2019
  ident: ref_2
  article-title: Prevalence of mental disorders in China: A cross-sectional epidemiological study
  publication-title: Lancet Psychiatry
  doi: 10.1016/S2215-0366(18)30511-X
– ident: ref_43
  doi: 10.18653/v1/2023.emnlp-main.370
– ident: ref_19
– volume: 3
  start-page: 146
  year: 2020
  ident: ref_12
  article-title: Rethinking domain adaptation for machine learning over clinical language
  publication-title: JAMIA Open
  doi: 10.1093/jamiaopen/ooaa010
– ident: ref_11
  doi: 10.18653/v1/2023.acl-long.555
– volume: 29
  start-page: 1930
  year: 2023
  ident: ref_15
  article-title: Large language models in medicine
  publication-title: Nat. Med.
  doi: 10.1038/s41591-023-02448-8
SSID ssj0001325264
Score 2.3008244
Snippet On social media, users often express their personal feelings, which may exhibit cognitive distortions or even suicidal tendencies on certain specific topics....
SourceID doaj
pubmedcentral
proquest
gale
pubmed
crossref
SourceType Open Website
Open Access Repository
Aggregation Database
Index Database
StartPage 882
SubjectTerms Algorithms
Annotations
Artificial intelligence
Behavior modification
Classification
cognitive distortions
Datasets
Deep learning
Digital media
Distortion
Emotions
Labeling
Language
large language model
Large language models
Learning
Machine learning
Mental depression
Mental disorders
Mental health
Performance evaluation
Professionals
Prompt engineering
Risk factors
Sentiment analysis
Social media
Social networks
Subject specialists
Suicidal behavior
Suicidal ideation
Suicide
suicide detection
Suicides & suicide attempts
Supervised learning
Therapy
SummonAdditionalLinks – databaseName: ProQuest Central
  dbid: BENPR
  link: http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV1Lb9QwEB7B9gKHijeBgoyExCnaxHHihAvq9qEKiQq1VOotsmOnjUqTZbP7S_jDzDjeRwTiEilPOZrXN7bnG4CPstaZtFyGvE51KESah6owPKw4jyqR6KRyjDffzrOzK_H1Or32E26931a59onOUZuuojnyKQb-pCC6tezL_FdIXaNoddW30HgIe-iC83QCe7OT8-8X21mWhKcY8ofS4ATz-6luOrtl-os5IqY856Oo5Mj7_3bROzFqvH9yJyCdPoF9jyTZ4SD6p_DAts_g8Q6_4HP4fbmaky_orWGeSPWGqRZPaP83Hoe5SkYN0X6yGSrs7b1a3PWsa9lA7sOGMiV2rJYY7pb9Z3a03m_Ejh3DiNNb99HLVVM1Bt-5aHr8RtMyas5te8uGGmBGi0LqBVydnvw4Ogt9F4awSjBvDoVFqdVUr5vURaRzXcWui5-1uo5jjQgkMqaQlYi0iGzNtbAmUrWRkeOvk8lLmLRda18Di1IjjcoriVYvrERXkCoi_0GYoBApqACmaymU84Fso8QkheRW_ltuAcxIWJuniSzbXegWN6W3vVLUlho_SBUbidmo0YmKFP6ByDC5ymMbwCcSdUkmjfKslK9MwEETOVZ5iJAGUW8RFwEcrLWh9Lbel1vNDODD5jZaKS29qNZ2K_dMRnBBygBeDcqzGbPArBJhahpAPlKr0U-N77TNrWMCR9iRJ3mUvfn_uN7CI05ti4nJtziAyXKxsu8QSy31e28wfwAv0yHk
  priority: 102
  providerName: ProQuest
Title Supervised Learning and Large Language Model Benchmarks on Mental Health Datasets: Cognitive Distortions and Suicidal Risks in Chinese Social Media
URI https://www.ncbi.nlm.nih.gov/pubmed/40868395
https://www.proquest.com/docview/3243982856
https://www.proquest.com/docview/3246413077
https://pubmed.ncbi.nlm.nih.gov/PMC12383806
https://doaj.org/article/4feca507a1d7448db3a0a0134618281e
Volume 12
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwrV3fj5QwEG70fNEH42_Rc1MTE5_IQlto8W25vfVi4sXcecm9kZaWO6Kyl2P3L_EfdqZlV4gmvvgCgQIpzEznG9r5hpB3sjG5dEzGrMlMLESmYl1YFteMJbXghtee8ebzaX5yIT5dZpejUl-4JizQA4cPNxeNQ9Z-qVMrIZSwhutEA24ROSBjlTocfcHnjYIp_3eFswxcfUgJ5hDXz027dr8Z_lIGSEkpNvFGnrT_z6F55Jum6yZHjmj1iDwcECRdhJ4_Jndc94Q8GPEKPiU_z7c3OAb0ztKBQPWK6g4OcN03bMM_SoqF0L7TEhT1-oe-_dbTdUcDqQ8N6Ul0qTfg5jb9B3q0W2dEl55ZxOurf-j5tq1bC_ectT08o-0oFuV2vaMh95fiZJB-Ri5Wx1-PTuKh-kJcc4iXY-FAWg3m6fKmSIwydeqr9zlnmjQ1gDwSawtZi8SIxDXMCGcT3ViZeN46yZ-Tg27duZeEJpmVVqtagrULJ2EIyDSS_gA80IAQdETmOylUN4Fko4LgBOVW_V1uESlRWPurkSTbnwDVqQbVqf6lOhF5j6Ku0JRBnrUeMhKg00iKVS0AygDaLdIiIoc7bagGG-8rgKK8QALAPCJv981gnTjloju33vprcoQJUkbkRVCefZ8FRJMAT7OIqIlaTV5q2tK1154BHOCG4irJX_2Pz_Ca3GdY1Bh5fotDcrC53bo3gLQ2ZkbuqtXHGbm3KJflCvbl8emXs5k3tV-PgCxz
linkProvider Directory of Open Access Journals
linkToHtml http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1bb9MwFLbG9gA8IO4EBhgJxFPUxHbiBAmhdd3Usa1Cu0h7C3bsbBGQlKYV4nfwP_iNnJNL2wjE214qpUksR-f2HdvnO4S8lpkOpWXSZVmgXSGCyFWxYW7KmJcKrnlaM94cT8Lxufh4EVxskN9dLQweq-x8Yu2oTZniGvkAAj-PkW4t_DD97mLXKNxd7VpoNGpxaH_-gJSten8wAvm-YWx_72x37LZdBdyUQx7oCguzyLD-lGexpyOd-nVXOmt15vsaIqpnTCxT4Wnh2YxpYY2nMiO9mo9Nchj3BtkCmBGDFW0N9yafTlarOpwFADGaUmTOY2-g89KumAV9BggtilgvCtbNAv4OCWsxsX9ecy0A7t8ld1rkSncaVbtHNmxxn9xe4zN8QH6dLqboeypraEvceklVARd43hx-m7VRig3YvtIhGMjVNzX7UtGyoA2ZEG3KouhIzSG8zqt3dLc730RHNaNJbSf1oKeLPM0NvHOSVzBGXlBsBm4rS5uaY4qbUOohOb8W-Twim0VZ2CeEeoGRRkWpBC8jrATXEygkGwJYogCZKIcMOikk04bcI4GkCOWW_FtuDhmisJZPIzl3_Uc5u0xaW09EZrHRhFS-kZD9Gs2Vp-ALRAjJXORbh7xFUSfoQkCeqWorIWDSSMaV7ACEApQd-7FDtjttSFrfUiUrS3DIq-Vt8Aq41aMKWy7qZ0KEJ1I65HGjPMs5C8hiARYHDol6atX7qP6dIr-qmccB5kQ88sKn_5_XS3JzfHZ8lBwdTA6fkVsMWyYji3C8TTbns4V9Djhurl-0xkPJ5-u21z8Qj125
linkToPdf http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1fa9RAEB_qFUQfxP9Gq66g-BQu2WyyiSDS6_VorR6ltdC3uJvdtEFNzssd4ufw2_jpnEly_1B868tBLsmyYXZmfrM78xuAlzLXkbRcujwPtStEGLsqMdzNOPcyEeggaxhvPo6jgzPx_jw834Lfi1oYSqtc2MTGUJsqoz3yPjr-ICG6taifd2kRx8PRu8l3lzpI0Unrop1Gu0SO7M8fGL7Vbw-HKOtXnI_2P-0duF2HATcLMCZ0hcUZ5VSLGuSJp2Od-U2HOmt17vsavatnTCIz4Wnh2ZxrYY2nciO9hptNBjjuNdjGd6K4B9uD_fHxyWqHJ-Ahwo22LDkIEq-vi8quWAZ9jmgtjvmGR2waB_ztHtb842bu5pozHN2GWx2KZbvtsrsDW7a8CzfXuA3vwa_T-YTsUG0N60hcL5gq8YJyz_G33Sdl1IztKxugslx-U9MvNatK1hILsbZEig3VDF3trH7D9ha5TmzYsJs0OtMMejovssLgOydFjWMUJaPG4La2rK0_ZnQgpe7D2ZXI5wH0yqq0j4B5oZFGxZlEiyOsRDMUKiIeQoiiEKUoB_oLKaSTlugjxQCJ5Jb-W24ODEhYy6eJqLv5o5pepJ3epyK31HRCKt9IjISNDpSn8AtEhIFd7FsHXpOoUzInKM9MdVUROGki5kp3EU4h4k78xIGdxWpIOztTpyutcODF8jZaCDr2UaWt5s0zEUEVKR142C6e5ZwFRrQIkUMH4o1ltfFRm3fK4rJhIUfIEwexFz3-_7yew3XU0_TD4fjoCdzg1D2ZCIWTHejNpnP7FCHdTD_rdIfB56tW1z8VdmHl
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Supervised+Learning+and+Large+Language+Model+Benchmarks+on+Mental+Health+Datasets%3A+Cognitive+Distortions+and+Suicidal+Risks+in+Chinese+Social+Media&rft.jtitle=Bioengineering+%28Basel%29&rft.au=Qi%2C+Hongzhi&rft.au=Fu%2C+Guanghui&rft.au=Li%2C+Jianqiang&rft.au=Song%2C+Changwei&rft.date=2025-08-19&rft.issn=2306-5354&rft.eissn=2306-5354&rft.volume=12&rft.issue=8&rft_id=info:doi/10.3390%2Fbioengineering12080882&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2306-5354&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2306-5354&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2306-5354&client=summon