Skip to main content
Peer-reviewed Authoritative

A Systematic Literature Review of the Use of Computational Text Analysis Methods in Intimate Partner Violence Research

PRISMA systematic review across eight databases of 22 studies applying computational text-analysis and NLP methods to intimate-partner-violence research, spanning rule-based, classical machine-learning, deep-learning, and topic-modelling approaches. Data sources were predominantly social-media text, plus police, health/social-care, and litigation texts.

Publisher

Journal of Family Violence (Springer)

Published

21 Mar 2023

Added

today

Key Findings

  • 22 studies applied computational text analysis to IPV research, most using social-media data (15 of 22)
  • Methods spanned rule-based, classical ML, deep learning, and topic modelling; evaluation used held-out/k-fold accuracy and F1
  • Identifies gaps including dataset scarcity, limited generalisability, and ethical/consent concerns in IPV text mining

Methodology Notes

Peer-reviewed, Journal of Family Violence (online ahead of print 21 March 2023), DOI 10.1007/s10896-023-00517-7. PRISMA-P systematic review (UCL authorship). Verified via the open-access PMC full text (PMC10028783).

Authors

Lilly Neubauer, Isabel Straw, Enrico Mariconti, Leonie Maria Tanczer

Tags

ipvsystematic-reviewtext-analysisnlpabuse

Cite This

APA

Lilly Neubauer et al. (2023). A Systematic Literature Review of the Use of Computational Text Analysis Methods in Intimate Partner Violence Research. Journal of Family Violence (Springer). https://link.springer.com/article/10.1007/s10896-023-00517-7