Tracking and analyzing dynamics of news-cycles during global pandemics

Abstract

How does the tone of reporting during a disease outbreak change in relation to the number of cases, categories of victims, and accumulating deaths? How do newspapers and medical journals contribute to the narrative of a historical pandemic? Can data mining experts help history scholars to scale up the process of examining articles, extracting new insights and understanding the public opinion of a pandemic? We explore these problems in this paper, using the 19thcentury Russian Flu epidemic as an example. We study two different types of historical data sources: the US medical discussion and popular reporting during the epidemic, from its outbreak in late 1889 through the successive waves that lasted through 1893. We analyze and compare these articles and reports to answer three major questions. First, we analyze how newspapers and medical journals report the Russian flu and describe the situation. Next, we help historians in understanding the tone of related reports and how they vary across data sources. We also examine the temporal changes in the discussion to get an in-depth understanding of how public opinion changed about the pandemic. Finally, we aggregate all of the algorithms in an easy to use framework GrippeStory to help history scholars investigate historical pandemic data in general, across chronological periods and locations. Our extensive experiments and analysis on a large number of historical articles show that GrippeStory gives meaningful and useful results for historians and it outperforms the baselines.