Evolving Face Mask Guidance During a Pandemic and Potential Harm to Public Perception: Infodemiology Study of Sentiment and Emotion on Twitter.

BACKGROUND: Throughout the COVID-19 pandemic, US Centers for Disease Control and Prevention policies on face mask use fluctuated. Understanding how public health communications evolve around key policy decisions may inform future decisions on preventative measures by aiding the design of communication strategies (eg, wording, timing, and channel) that ensure rapid dissemination and maximize both widespread adoption and sustained adherence. OBJECTIVE: We aimed to assess how sentiment on masks evolved surrounding 2 changes to mask guidelines: (1) the recommendation for mask use on April 3, 2020, and (2) the relaxation of mask use on May 13, 2021. METHODS: We applied an interrupted time series method to US Twitter data surrounding each guideline change. Outcomes were changes in the (1) proportion of positive, negative, and neutral tweets and (2) number of words within a tweet tagged with a given emotion (eg, trust). Results were compared to COVID-19 Twitter data without mask keywords for the same period. RESULTS: There were fewer neutral mask-related tweets in 2020 (ß=-3.94 percentage points, 95% CI -4.68 to -3.21; P<.001) and 2021 (ß=-8.74, 95% CI -9.31 to -8.17; P<.001). Following the April 3 recommendation (ß=.51, 95% CI .43-.59; P<.001) and May 13 relaxation (ß=3.43, 95% CI 1.61-5.26; P<.001), the percent of negative mask-related tweets increased. The quantity of trust-related terms decreased following the policy change on April 3 (ß=-.004, 95% CI -.004 to -.003; P<.001) and May 13 (ß=-.001, 95% CI -.002 to 0; P=.008). CONCLUSIONS: The US Twitter population responded negatively and with less trust following guideline shifts related to masking, regardless of whether the guidelines recommended or relaxed mask usage. Federal agencies should ensure that changes in public health recommendations are communicated concisely and rapidly.

The Relationship between Sextortion during COVID-19 and Pre-pandemic Intimate Partner Violence: A Large Study of Victimization among Diverse U.S Men and Women

In a large and diverse sample of U. S. adults, we assessed participants' experience with pre-COVID in-person intimate partner violence (IPV) victimization and with sextortion victimization during COVID to better understand the relationship between these phenomena. Experiencing sexual IPV pre-COVID increased the likelihood that men and women would experience sextortion during COVID. Men, Black and Native women, LGBTQ individuals, and emerging adults more often experienced sextortion during COVID than other groups. Implications for research on technology-facilitated sexual violence and practice with survivors are explored.

Evaluating criminal justice reform during COVID-19: The need for a novel sentiment analysis package.

The health and safety of incarcerated persons and correctional personnel have been prominent in the U.S. news media discourse during the COVID-19 pandemic. Examining changing attitudes toward the health of the incarcerated population is imperative to better assess the extent to which the general public favors criminal justice reform. However, existing natural language processing lexicons that underlie current sentiment analysis (SA) algorithms may not perform adequately on news articles related to criminal justice due to contextual complexities. News discourse during the pandemic has highlighted the need for a novel SA lexicon and algorithm (i.e., an SA package) tailored for examining public health policy in the context of the criminal justice system. We analyzed the performance of existing SA packages on a corpus of news articles at the intersection of COVID-19 and criminal justice collected from state-level outlets between January and May 2020. Our results demonstrated that sentence sentiment scores provided by three popular SA packages can 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.

Machine Learning Maps Research Needs in COVID-19 Literature.

As of August 2020, thousands of COVID-19 (coronavirus disease 2019) publications have been produced. Manual assessment of their scope is an overwhelming task, and shortcuts through metadata analysis (e.g., keywords) assume that studies are properly tagged. However, machine learning approaches can rapidly survey the actual text of publication abstracts to identify research overlap between COVID-19 and other coronaviruses, research hotspots, and areas warranting exploration. We propose a fast, scalable, and reusable framework to parse novel disease literature. When applied to the COVID-19 Open Research Dataset, dimensionality reduction suggests that COVID-19 studies to date are primarily clinical, modeling, or field based, in contrast to the vast quantity of laboratory-driven research for other (non-COVID-19) coronavirus diseases. Furthermore, topic modeling indicates that COVID-19 publications have focused on public health, outbreak reporting, clinical care, and testing for coronaviruses, as opposed to the more limited number focused on basic microbiology, including pathogenesis and transmission.