RESUMEN
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.
RESUMEN
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.
RESUMEN
BACKGROUND: This paper combines complementary attributes of web and general population surveys to estimate cannabis consumption and spending in Washington State. It compares those estimates to legal sales recorded by the state's seed-to-sale tracking system, and thus exploits a rare opportunity to contrast two independent estimates for the same cannabis market. This sheds light on the question of whether nontrivial amounts of black market sales continue even after a state allows licensed production and sale. METHODS: Prevalence of past-month use is estimated from the 2015/16 U.S. National Survey on Drug Use and Health, adjusted for under-reporting. Estimates of consumption and spending per user broken down by age, gender, and frequency of use are developed from RAND's 2013 survey of cannabis users in Washington State. Supply side estimates come from the Washington State Liquor and Cannabis Board's seed-to-sale tracking system. They are expressed in terms of spending, equivalent-weight of flowers, and THC, with THC for edibles imputed using a machine learning technique called random forests. RESULTS: For the period July 1, 2016 to June 30, 2017, Washington's seed-to-sale data record sales from licensed cannabis stores of $1.17B and across all products an amount of THC that is equivalent to roughly 120-150â¯MT of flower. Survey responses suggest that amounts spent and quantities consumed are larger than that, perhaps on the order of $1.66B and over 200â¯MT, respectively. CONCLUSION: A perfect match is not expected because of sales to tourists, residual black market activity, production for medical purposes, and diversion across state lines. Nonetheless, the results suggest that three years after state-licensed stores opened, there remained considerable consumption of cannabis supplied outside of the licensed system.
