TY - JOUR
T1 - Using machine learning to assess rape reports
T2 - Sentiment analysis detection of officers' “signaling” about victims' credibility
AU - Lovell, Rachel E.
AU - Klingenstein, Joanna
AU - Du, Jiaxin
AU - Overman, Laura
AU - Sabo, Danielle
AU - Ye, Xinyue
AU - Flannery, Daniel J.
N1 - Funding Information:
We want to acknowledge the survivors. These are not data points but individuals who endured intense, intimate trauma—survivors. To these survivors: We hold your stories with the reverence and respect they deserve. Thank you for inadvertently sharing them with us. We aim to leverage your stories to inform and improve our response to rape. We would like to acknowledge and thank the Cleveland Division of Police and Cuyahoga County Prosecutor's Office for their support and the researchers who worked on this project but are not co-authors, including Misty Luminais, Margaret McGuire, Grayson Holt, Julia Zalewski, and Madeline Myers.
Funding Information:
This project was supported by Grant No. 2018-VA-CX-0002, awarded by the National Institute of Justice (NIJ). NIJ is the research, development and evaluation agency of the United States Department of Justice, Office of Justice Programs (OJP). Points of view or opinions in this document are those of the authors and do not necessarily represent the official position or policies of the US Department of Justice.
Publisher Copyright:
© 2023 Elsevier Ltd
PY - 2023/9/1
Y1 - 2023/9/1
N2 - Purpose: The first of two articles from a larger study whose aim was to teach a computer to detect innuendo (or signaling) about a victim's credibility in incident reports of rape. This study explored the degree of sentiment and subjectivity in the reports and whether these predicted case progression and outcomes. Methods: We employed machine learning, specifically sentiment analysis to assess sentiment (opinion) and subjectivity of textual content. The sample consists of 5638 incident reports of rape with a sexual assault kit from a U.S., urban Midwestern jurisdiction. Results: Sentiment was detected, tended to skew near neutral/slightly negative and more subjective, and predicted case progression and outcomes, but was not quite what was expected. Conclusions: We taught a computer to detect signaling via tone that predicted case progression and outcomes. Findings indicate that the cases recommended for prosecution were longer and had positive sentiment and positive subjectivity. Cases not recommended for prosecution were shorter with more neutral statements of “fact” or observations. Implications and recommendations for improved, less biased report writing are provided.
AB - Purpose: The first of two articles from a larger study whose aim was to teach a computer to detect innuendo (or signaling) about a victim's credibility in incident reports of rape. This study explored the degree of sentiment and subjectivity in the reports and whether these predicted case progression and outcomes. Methods: We employed machine learning, specifically sentiment analysis to assess sentiment (opinion) and subjectivity of textual content. The sample consists of 5638 incident reports of rape with a sexual assault kit from a U.S., urban Midwestern jurisdiction. Results: Sentiment was detected, tended to skew near neutral/slightly negative and more subjective, and predicted case progression and outcomes, but was not quite what was expected. Conclusions: We taught a computer to detect signaling via tone that predicted case progression and outcomes. Findings indicate that the cases recommended for prosecution were longer and had positive sentiment and positive subjectivity. Cases not recommended for prosecution were shorter with more neutral statements of “fact” or observations. Implications and recommendations for improved, less biased report writing are provided.
KW - Attrition
KW - Natural language processing
KW - Sentiment analysis
KW - Sexual assault, machine learning
KW - Signaling
KW - Victim credibility
UR - http://www.scopus.com/inward/record.url?scp=85169065471&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85169065471&partnerID=8YFLogxK
U2 - 10.1016/j.jcrimjus.2023.102106
DO - 10.1016/j.jcrimjus.2023.102106
M3 - Article
AN - SCOPUS:85169065471
SN - 0047-2352
VL - 88
JO - Journal of Criminal Justice
JF - Journal of Criminal Justice
M1 - 102106
ER -