Google Trends (GT) is the most popular Big Data surveillance tool that helps researchers analyze temporal and geographical trends in online search terms or topics ( Mavragani and Ochoa, 2018b Mavragani et al., 2018a). Some of these studies have been used explicitly for the monitoring and forecasting of epidemics, such as Zika ( Farhadloo et al., 2018), Ebola ( Van Lent et al., 2017) and influenza ( Lu et al., 2018). Many infodemiologic studies have demonstrated the usefulness of real-time data in health assessment ( Van Lent et al., 2017 Wongkoblap et al., 2017 Farhadloo et al., 2018 Lu et al., 2018 Mavragani et al., 2018b Xu et al., 2020). Data used for infodemiology, which may or may not have been intended for epidemiological purposes, can be retrieved from Twitter tweets, Facebook posts, or Google search queries. Digital epidemiology, otherwise known as infodemiology, uses digital data or online sources to gain insight into disease dynamics and inform public health policies ( Eysenbach, 2009 Salathé, 2018). The outbreak of many infectious diseases in the current digital age, including the coronavirus, has led to significant interest in using digital epidemiology and big data tools to enhance disease surveillance and modeling. Since then, the Nigerian COVID-19 cases grew steadily to more than 250,000 cases and 3,000 deaths in February 2022 ( Worldometer, 2022). Nigeria, in particular, reported an index case of COVID-19 on February 27, 2020, making it the first in West Africa ( NCDC, 2020). The coronavirus pandemic did not spare the African continent-cases have already been reported in all 54 African countries. The COVID-19 pandemic has led to enormous social and economic harm worldwide, including job loss, severe illness, and death ( Pan et al., 2020). ![]() As of February 2022, more than 434 million cases and over 5.9 million fatalities have been documented worldwide, according to the John Hopkins University ( Dong et al., 2020). It has had more global and rapid spread since the first confirmed cases in China in December 2019. ![]() The coronavirus (COVID-19) pandemic has been arguably the most critical public health challenge of the 21st century. Google trends data enhanced the predictive ability of a traditionally based model and should be considered a suitable method to enhance infectious disease modeling. The difference in predictive performances was significant when using a two-sided Diebold-Mariano test (DM = 6.75, p < 0.001) for the 13 weeks. Corrected Akaike Information Criteria also favored the GT expanded model (869.4 vs. The GT expanded model achieved better forecasting accuracy (RMSE: 388.7 and MAE = 340.1). Predictions of the ARIMA model using solely reported case numbers resulted in an RMSE (root mean squared error) of 411.4 and mean absolute error (MAE) of 354.9. Preliminary results of contemporaneous correlations between COVID-related search terms and weekly COVID cases reveal “loss of smell,” “loss of taste,” “fever” (in order of magnitude) as significantly associated with the official cases. Statistical significance of the difference in predictions was determined with the two-sided Diebold-Mariano test. Forecast accuracies were compared visually and using RMSE (root mean square error) and MAE (mean average error). Model forecasts, both with and without GTD, were compared with weekly cases in the test set over 13 weeks. The utilized Google Trends (GT) variable was added to the ARIMA model as a regressor. Several COVID-related search terms were theoretically and empirically assessed for initial screening. ARIMA models were fitted to describe reported weekly COVID cases using the training set. The reported weekly incidence numbers and the GT data were split into training and testing sets. Data on the Nigerian weekly COVID-19 cases spanning through March 1, 2020, to May 31, 2021, were matched with internet search data from Google Trends. We used Google Trends data to track COVID-19 incidences and assessed whether they could complement traditional data based solely on reported case numbers. It is of interest to verify the utility of these methods using a Nigerian case study. Infodemiologic methods could be used to enhance modeling infectious diseases. ![]() Centre for Applied Data Science, College of Business and Economics, University of Johannesburg, Johannesburg, South Africa.Lateef Babatunde Amusa *, Hossana Twinomurinzi and Chinedu Wilfred Okonkwo
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |