Evaluating Criminal Justice Reform During COVID-19: The Need for a Novel Sentiment Analysis Package

14 Pages Posted: 3 Mar 2021

See all articles by Divya Ramjee

Divya Ramjee

American University (Washington, DC)

Louisa Smith

Harvard University - Harvard TH Chan School of Public Health

Anhvinh Doanvo

Independent

Marie Charpignon

Massachusetts Institute of Technology (MIT)

Alyssa McNulty

Texas A&M University

Angel Desai

International Society for Infectious Diseases

Maimuna S. Majumder

Boston Children's Hospital - Computational Health Informatics Program; Harvard University - Harvard Medical School

Date Written: February 27, 2021

Abstract

Existing natural language processing lexicons that underlie current sentiment analysis (SA) algorithms may not perform adequately in certain academic disciplines depending on contextual complexities. The health and safety of incarcerated persons and correctional personnel have been prominent in news media discourse during the COVID-19 pandemic, potentially highlighting the need for a novel SA lexicon and algorithm that is tailored for the examination of public health policy in the context of the criminal justice system. We utilized a text corpus consisting of news articles at the intersection of COVID-19 and criminal justice to analyze the performance of existing lexicons collected across state-level outlets between January and May 2020. Our results demonstrated that sentence sentiment scores provided by three popular SA packages differ considerably from manually-curated ratings. This dissimilarity was especially pronounced when the text was more polarized, whether negatively or positively. A randomly selected set of 1,000 manually scored sentences, and the corresponding binary document term matrices, were used to train two new sentiment prediction algorithms (i.e., linear regression and random forest regression) to verify the performance of the manually-curated ratings. By better accounting for the unique context in which incarceration-related terminologies are used in news media, both of our proposed models outperformed all existing SA packages considered for comparison. Our findings suggest that there is a need to develop a novel lexicon, and potentially an accompanying algorithm, for analysis of text related to public health within the criminal justice system, as well as criminal justice more broadly.

Keywords: COVID-19, criminal justice, sentiment analysis, text analysis, public health, health, safety, public policy, lexicon, algorithm, NLP

Suggested Citation

Ramjee, Divya and Smith, Louisa and Doanvo, Anhvinh and Charpignon, Marie and McNulty, Alyssa and Desai, Angel and Majumder, Maimuna, Evaluating Criminal Justice Reform During COVID-19: The Need for a Novel Sentiment Analysis Package (February 27, 2021). Available at SSRN: https://ssrn.com/abstract=3792638 or http://dx.doi.org/10.2139/ssrn.3792638

Divya Ramjee (Contact Author)

American University (Washington, DC) ( email )

4400 Massachusetts Ave, NW
Washington, DC 20016
United States

Louisa Smith

Harvard University - Harvard TH Chan School of Public Health ( email )

Bostone, MA 02115
United States

Anhvinh Doanvo

Independent ( email )

Marie Charpignon

Massachusetts Institute of Technology (MIT) ( email )

77 Massachusetts Avenue
50 Memorial Drive
Cambridge, MA 02139-4307
United States

Alyssa McNulty

Texas A&M University ( email )

Langford Building A
798 Ross St.
College Station, TX 77843-3137
United States

Angel Desai

International Society for Infectious Diseases ( email )

Brookline, MA

Maimuna Majumder

Boston Children's Hospital - Computational Health Informatics Program ( email )

United States

Harvard University - Harvard Medical School ( email )

25 Shattuck St
Boston, MA 02115
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
101
Abstract Views
983
rank
319,967
PlumX Metrics