Using Text and Data Mining to Gain Insights on COVID-19
Government agencies have taken various regulatory actions in response to the COVID-19 pandemic. What are people saying about these regulatory responses? Which ones have prompted the most discussion? And, most importantly, which regulations should be most urgently addressed?
To answer these questions, George Washington University researcher Zhoudan Xie collaborated with ProQuest on a special text and data mining (TDM) project. Using TDM Studio, ProQuest’s new text and data mining solution, she analyzed a massive set of recent U.S. news articles related to both COVID-19 and regulation. Her goal: to see what topics were written about most frequently.
From an initial idea to the final output, TDM Studio gives researchers like Xie access to current news and scholarly content in record time by significantly reducing the time and effort needed for the up-front data collection and formatting. This enables researchers to quickly reveal relationships, patterns and connections within and between datasets from a variety of sources, including current and historical ProQuest content.
Using a dataset of a million-plus articles specifically curated for COVID-19 research, Xie’s analysis identified 16 topics in articles between January and April 2020 – from “quarantine and reopening” to “oil price.” These topics demonstrate different levels of prevalence and trends over time, reflecting changes in media attention on regulatory policy response to COVID-19.
Nobody can read a million articles manually, so a tool like TDM Studio is crucial for analyzing this volume of data. The project helped Xie understand the general patterns and trends in regulatory issues being talked about in the news related to COVID-19.
“With the rich resources provided by TDM Studio,” said Xie, “further text mining and analysis can be applied to answer more specific questions in regulatory studies, generating insights that may inform policymakers and help the economy recover from the pandemic.”