Quantitative Methods for Analyzing Intimate Partner Violence in Microblogs: Observational Study.
intimate partner violence
natural language processing
social media
Journal
Journal of medical Internet research
ISSN: 1438-8871
Titre abrégé: J Med Internet Res
Pays: Canada
ID NLM: 100959882
Informations de publication
Date de publication:
19 11 2020
19 11 2020
Historique:
received:
02
07
2019
accepted:
26
07
2020
revised:
20
12
2019
entrez:
19
11
2020
pubmed:
20
11
2020
medline:
2
3
2021
Statut:
epublish
Résumé
Social media is a rich, virtually untapped source of data on the dynamics of intimate partner violence, one that is both global in scale and intimate in detail. The aim of this study is to use machine learning and other computational methods to analyze social media data for the reasons victims give for staying in or leaving abusive relationships. Human annotation, part-of-speech tagging, and machine learning predictive models, including support vector machines, were used on a Twitter data set of 8767 #WhyIStayed and #WhyILeft tweets each. Our methods explored whether we can analyze micronarratives that include details about victims, abusers, and other stakeholders, the actions that constitute abuse, and how the stakeholders respond. Our findings are consistent across various machine learning methods, which correspond to observations in the clinical literature, and affirm the relevance of natural language processing and machine learning for exploring issues of societal importance in social media.
Sections du résumé
BACKGROUND
Social media is a rich, virtually untapped source of data on the dynamics of intimate partner violence, one that is both global in scale and intimate in detail.
OBJECTIVE
The aim of this study is to use machine learning and other computational methods to analyze social media data for the reasons victims give for staying in or leaving abusive relationships.
METHODS
Human annotation, part-of-speech tagging, and machine learning predictive models, including support vector machines, were used on a Twitter data set of 8767 #WhyIStayed and #WhyILeft tweets each.
RESULTS
Our methods explored whether we can analyze micronarratives that include details about victims, abusers, and other stakeholders, the actions that constitute abuse, and how the stakeholders respond.
CONCLUSIONS
Our findings are consistent across various machine learning methods, which correspond to observations in the clinical literature, and affirm the relevance of natural language processing and machine learning for exploring issues of societal importance in social media.
Identifiants
pubmed: 33211021
pii: v22i11e15347
doi: 10.2196/15347
pmc: PMC7714648
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Research Support, U.S. Gov't, Non-P.H.S.
Langues
eng
Sous-ensembles de citation
IM
Pagination
e15347Informations de copyright
©Christopher Michael Homan, J Nicolas Schrading, Raymond W Ptucha, Catherine Cerulli, Cecilia Ovesdotter Alm. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 19.11.2020.
Références
Am J Community Psychol. 2005 Sep;36(1-2):71-84
pubmed: 16134045
Lancet. 2002 Oct 5;360(9339):1083-8
pubmed: 12384003
Womens Health Issues. 2011 Jul-Aug;21(4 Suppl):S190-4
pubmed: 21724140
J Gen Intern Med. 2011 Aug;26(8):930-3
pubmed: 21448753
Fam Syst Health. 2014 Sep;32(3):259-270
pubmed: 24911769
PLoS One. 2011;6(12):e26752
pubmed: 22163266
Online Soc Netw Media. 2017 Aug;2:32-44
pubmed: 29278258
Artif Life. 2011 Summer;17(3):237-51
pubmed: 21554117
J Womens Health (Larchmt). 2009 May;18(5):625-31
pubmed: 19445614
Artif Intell Med. 2014 Jul;61(3):153-63
pubmed: 24613716
IEEE Trans Neural Netw Learn Syst. 2017 Oct;28(10):2222-2232
pubmed: 27411231
Proc Natl Acad Sci U S A. 2014 Jun 17;111(24):8788-90
pubmed: 24889601
Am J Public Health. 2012 May;102 Suppl 2:S195-200
pubmed: 22401525
Violence Against Women. 1998 Jun;4(3):262-90
pubmed: 12296014
Science. 2011 Sep 30;333(6051):1878-81
pubmed: 21960633
J Womens Health (Larchmt). 2011 Dec;20(12):1923-9
pubmed: 21988551
J Fam Violence. 2012 Nov;27(8):773-781
pubmed: 23226694
Popul Rep L. 1999 Dec;(11):1-43
pubmed: 11056940