Detecting Suicidal Ideation on Forums: Proof-of-Concept Study

被引:69
|
作者
Emre Aladag, Ahmet [1 ,2 ]
Muderrisoglu, Serra [3 ]
Akbas, Naz Berfu [4 ]
Zahmacioglu, Oguzhan [5 ]
Bingol, Haluk O. [1 ]
机构
[1] Bogazici Univ, Dept Comp Engn, TR-34342 Istanbul, Turkey
[2] Amazon Res, Madrid, Spain
[3] Bogazici Univ, Dept Psychol, Istanbul, Turkey
[4] Yeditepe Univ, Dept Psychiat, Med Sch, Istanbul, Turkey
[5] Yeditepe Univ, Med Sch, Dept Child & Adolescent Psychiat, Istanbul, Turkey
关键词
suicide; suicidal ideation; suicidality; detection; prevention; classification model; text mining; machine learning; artificial intelligence; suicidal surveillance; SOCIAL MEDIA;
D O I
10.2196/jmir.9840
中图分类号
R19 [保健组织与事业(卫生事业管理)];
学科分类号
摘要
Background: In 2016, 44,965 people in the United States died by suicide. It is common to see people with suicidal ideation seek help or leave suicide notes on social media before attempting suicide. Many prefer to express their feelings with longer passages on forums such as Reddit and blogs. Because these expressive posts follow regular language patterns, potential suicide attempts can be prevented by detecting suicidal posts as they are written. Objective: This study aims to build a classifier that differentiates suicidal and nonsuicidal forum posts via text mining methods applied on post titles and bodies. Methods: A total of 508,398 Reddit posts longer than 100 characters and posted between 2008 and 2016 on SuicideWatch, Depression, Anxiety, and ShowerThoughts subreddits were downloaded from the publicly available Reddit dataset. Of these, 10,785 posts were randomly selected and 785 were manually annotated as suicidal or nonsuicidal. Features were extracted using term frequency-inverse document frequency, linguistic inquiry and word count, and sentiment analysis on post titles and bodies. Logistic regression, random forest, and support vector machine (SVM) classification algorithms were applied on resulting corpus and prediction performance is evaluated. Results: The logistic regression and SVM classifiers correctly identified suicidality of posts with 80% to 92% accuracy and F1 score, respectively, depending on different data compositions closely followed by random forest, compared to baseline ZeroR algorithm achieving 50% accuracy and 66% F1 score. Conclusions: This study demonstrated that it is possible to detect people with suicidal ideation on online forums with high accuracy. The logistic regression classifier in this study can potentially be embedded on blogs and forums to make the decision to offer real-time online counseling in case a suicidal post is being written.
引用
收藏
页数:10
相关论文
共 50 条
  • [1] Remote blended treatment for individuals with suicidal ideation: A single-arm proof-of-concept trial
    Buescher, Rebekka
    Teismann, Tobias
    Hartleitner, Paula
    Klein, Jan Philipp
    Baumeister, Harald
    Sander, Lasse B.
    [J]. CLINICAL PSYCHOLOGY & PSYCHOTHERAPY, 2024, 31 (01)
  • [2] A proof-of-concept study for detecting ctDNA in rare thyroid cancers
    Kuri, D. M. Vodopivec
    Chen, Y. H.
    Cabanillas, M. E.
    Hu, M. I.
    Busaidy, N. L.
    Dadu, R.
    Budde, G.
    Malhotra, M.
    Diersing, J.
    Zafereo, M. E.
    [J]. ANNALS OF ONCOLOGY, 2022, 33 (07) : S1296 - S1296
  • [3] Brief CBT for insomnia delivered in primary care to patients endorsing suicidal ideation: a proof-of-concept randomized clinical trial
    Pigeon, Wilfred R.
    Funderburk, Jennifer S.
    Cross, Wendi
    Bishop, Todd M.
    Crean, Hugh F.
    [J]. TRANSLATIONAL BEHAVIORAL MEDICINE, 2019, 9 (06) : 1169 - 1177
  • [4] Just a Small, Proof-of-Concept Study
    Novack, Gary D.
    [J]. OCULAR SURFACE, 2009, 7 (02): : 111 - 112
  • [5] Phonetic compliance: a proof-of-concept study
    Delvaux, Veronique
    Huet, Kathy
    Piccaluga, Myriam
    Harmegnies, Bernard
    [J]. FRONTIERS IN PSYCHOLOGY, 2014, 5
  • [6] A PROOF-OF-CONCEPT STUDY OF A MAGNETORHEOLOGICAL MICROPUMP
    Cesmeci, Sevki
    Hassan, Rubayet
    Thompson, Mark
    [J]. PROCEEDINGS OF ASME 2022 INTERNATIONAL MECHANICAL ENGINEERING CONGRESS AND EXPOSITION, IMECE2022, VOL 2B, 2022,
  • [7] Inhaled insulin: A proof-of-concept study
    Cefalu, WT
    [J]. ANNALS OF INTERNAL MEDICINE, 2001, 134 (09) : 795 - 795
  • [8] Radiomics for detecting prostate cancer bone metastases invisible in CT: a proof-of-concept study
    Hinzpeter, Ricarda
    Baumann, Livia
    Guggenberger, Roman
    Huellner, Martin
    Alkadhi, Hatem
    Baessler, Bettina
    [J]. EUROPEAN RADIOLOGY, 2022, 32 (03) : 1823 - 1832
  • [9] Radiomics for detecting prostate cancer bone metastases invisible in CT: a proof-of-concept study
    Ricarda Hinzpeter
    Livia Baumann
    Roman Guggenberger
    Martin Huellner
    Hatem Alkadhi
    Bettina Baessler
    [J]. European Radiology, 2022, 32 : 1823 - 1832
  • [10] Machine Learning Decision Support for Detecting Lipohypertrophy With Bedside Ultrasound: Proof-of-Concept Study
    Bandari, Ela
    Beuzen, Tomas
    Habashy, Lara
    Raza, Javairia
    Yang, Xudong
    Kapeluto, Jordanna
    Meneilly, Graydon
    Madden, Kenneth
    [J]. JMIR FORMATIVE RESEARCH, 2022, 6 (05)