TY - JOUR
T1 - Next generation media monitoring
T2 - Global coverage of electronic nicotine delivery systems (electronic cigarettes) on Bing, Google and Twitter, 2013-2018
AU - Ayers, John W.
AU - Dredze, Mark
AU - Leas, Eric C.
AU - Caputi, Theodore L.
AU - Allem, Jon Patrick
AU - Cohen, Joanna E.
N1 - Funding Information:
This work was supported by an award from the Institute for Global Tobacco Control at the Johns Hopkins Bloomberg School of Public Health, with funding from the The Bloomberg Initiative to Reduce Tobacco Use. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funder. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. This work was supported by an award to the Institute for Global Tobacco Control at the Johns Hopkins Bloomberg School of Public Health, with funding from The Bloomberg Initiative to Reduce Tobacco Use. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funder. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Publisher Copyright:
© 2018 Ayers et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
PY - 2018/11
Y1 - 2018/11
N2 - News media monitoring is an important scientific tool. By treating news reporters as data collectors and their reports as qualitative accounts of a fast changing public health landscape, researchers can glean many valuable insights. Yet, there have been surprisingly few innovations in public health media monitoring, with nearly all studies relying on labor-intensive content analyses limited to a small number of media reports. We propose to advance this subfield by using scalable machine learning. In potentially the largest contemporary public health media monitoring study to date, we systematically characterize global news reports surrounding electronic cigarettes or electronic nicotine delivery systems (ENDS) using natural language processing techniques. News reports including ENDS terms (e.g., “electronic cigarettes”) from over 100,000 sources (all sources archived on Google News or Bing News, as well as all news articles shared on Twitter) were monitored for 1 January 2013 through 31 July 2018. The geographic and subject (e.g., prevalence, bans, quitting, warnings, marketing, prices, age, flavor and industry) foci of news articles, their popularity among readers who share news on social media, and the sentiment behind news articles were assessed algorithmically. Globally there were 86,872 ENDS news reports with coverage increasing from 8 (standard deviation [SD] = 8) stories per day in 2013 to 75 (SD = 56) stories per day during 2018. The focus of ENDS news spanned 148 nations, with the plurality focusing on the United States (34% of all news). Potentially overlooked hotspots of ENDS media activity included China, Egypt, Russia, Ukraine, and Paraguay. The most common subject was warnings about ENDS (18%), followed by bans on using ENDS (13%) and ENDS prices (9%). Flavor and age restrictions were the least covered news subjects (~1% each). Among different subject foci, reports on quitting cigarettes using ENDS had the highest probability of scoring in the top three deciles of popularity rankings. Moreover, ENDS news on quitting and prices had a more positive sentiment on average than news with other subject foci. Public health leaders can use these trends to stay abreast of how ENDS are portrayed in the media, and potentially how the public perceives ENDS. Because our analytical strategies are updated in near real time, we aim to make media monitoring part of standard practice to support evidence-based tobacco control in the future.
AB - News media monitoring is an important scientific tool. By treating news reporters as data collectors and their reports as qualitative accounts of a fast changing public health landscape, researchers can glean many valuable insights. Yet, there have been surprisingly few innovations in public health media monitoring, with nearly all studies relying on labor-intensive content analyses limited to a small number of media reports. We propose to advance this subfield by using scalable machine learning. In potentially the largest contemporary public health media monitoring study to date, we systematically characterize global news reports surrounding electronic cigarettes or electronic nicotine delivery systems (ENDS) using natural language processing techniques. News reports including ENDS terms (e.g., “electronic cigarettes”) from over 100,000 sources (all sources archived on Google News or Bing News, as well as all news articles shared on Twitter) were monitored for 1 January 2013 through 31 July 2018. The geographic and subject (e.g., prevalence, bans, quitting, warnings, marketing, prices, age, flavor and industry) foci of news articles, their popularity among readers who share news on social media, and the sentiment behind news articles were assessed algorithmically. Globally there were 86,872 ENDS news reports with coverage increasing from 8 (standard deviation [SD] = 8) stories per day in 2013 to 75 (SD = 56) stories per day during 2018. The focus of ENDS news spanned 148 nations, with the plurality focusing on the United States (34% of all news). Potentially overlooked hotspots of ENDS media activity included China, Egypt, Russia, Ukraine, and Paraguay. The most common subject was warnings about ENDS (18%), followed by bans on using ENDS (13%) and ENDS prices (9%). Flavor and age restrictions were the least covered news subjects (~1% each). Among different subject foci, reports on quitting cigarettes using ENDS had the highest probability of scoring in the top three deciles of popularity rankings. Moreover, ENDS news on quitting and prices had a more positive sentiment on average than news with other subject foci. Public health leaders can use these trends to stay abreast of how ENDS are portrayed in the media, and potentially how the public perceives ENDS. Because our analytical strategies are updated in near real time, we aim to make media monitoring part of standard practice to support evidence-based tobacco control in the future.
UR - http://www.scopus.com/inward/record.url?scp=85056063126&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85056063126&partnerID=8YFLogxK
U2 - 10.1371/journal.pone.0205822
DO - 10.1371/journal.pone.0205822
M3 - Article
C2 - 30388126
AN - SCOPUS:85056063126
SN - 1932-6203
VL - 13
JO - PloS one
JF - PloS one
IS - 11
M1 - e0205822
ER -