Abstract
Real-time online data sources have contributed to timely and accurate forecasting of influenza activities while also suffered from instability and linguistic noise. Few previous studies have focused on unofficial online news articles, which are abundant in their numbers, rich in information, and relatively low in noise. This study examined whether monitoring both official and unofficial online news articles can improve influenza activity forecasting accuracy during influenza outbreaks. Data were retrieved from a Chinese commercial online platform and the website of the Chinese National Influenza Center. We modeled weekly fractions of influenza-related online news articles and compared them against weekly influenza-like illness (ILI) rates using autoregression analyses. We retrieved 153,958,695 and 149,822,871 online news articles focusing on the south and north of mainland China separately from 6 October 2019 to 17 May 2020. Our model based on online news articles could significantly improve the forecasting accuracy, compared to other influenza surveillance models based on historical ILI rates (p = 0.002 in the south; p = 0.000 in the north) or adding microblog data as an exogenous input (p = 0.029 in the south; p = 0.000 in the north). Our finding also showed that influenza forecasting based on online news articles could be 1–2 weeks ahead of official ILI surveillance reports. The results revealed that monitoring online news articles could supplement traditional influenza surveillance systems, improve resource allocation, and offer models for surveillance of other emerging diseases.
Highlights
Influenza is a severe public health concern internationally
Close inspection in the south suggests that adding microblogs as exogenous input (AR + Mblog) can decrease the RMSE = root-mean-squared error; R2 = coefficient of determination; MAE = mean absolute error
This study showed that unofficial online news articles accounted for the major part of present online news articles
Summary
Influenza is a severe public health concern internationally. Every year, influenza is estimated to cause 1 billion cases, 3–5 million severe cases, and 290,000–650,000 influenzarelated respiratory deaths worldwide [1,2] and cause an estimated 88,100 influenza-related excess respiratory deaths in China [3]. Through monitoring and evaluating influenza transmission, real-time influenza surveillance can contribute to early response and preparation for influenza epidemics [4]. The accuracy of real-time influenza activity forecasting strongly influences the effectiveness of the response and preparation [5]. The commonly used and well-established method to track influenza activity is through influenza-like illness (ILI). Reporting from sentinel hospitals all over the country [4,6]. The official ILI reports suffer from a delay of one to three weeks due to processing ILI-related information from sentinel hospitals [5,7]. Traditional influenza epidemics forecasting methods based on historical ILI reports cannot capture the influenza activity in a timely manner and sometimes
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
