Prediction of the Number of New Cases of 2019 Novel Coronavirus (COVID-19) Using a Social Media Search Index

Abstract

Purpose: Predicting number of novel Coronavirus Disease 2019 (COVID-19) new-suspected or confirmed cases is crucial in prevention and control of COVID-19 outbreak. Methods: Social media search indexes (SMSI) for dry cough, fever, chest distress, coronavirus, and pneumonia were collected from December 31, 2019, to February 9, 2020. The new-suspected cases of COVID-19 data were collected from January 20, 2020, to February 9, 2020. We used lagged series of SMSI to predict new-suspected COVID-19 case numbers during this period. To avoid overfitting, five methods, namely subset selection, forward selection, lasso regression, ridge regression, and elastic net, were used to estimate coefficients. We selected optimal method to predict new-suspected COVID-19 case numbers over subsequent 20 days. We further validated optimal method for new-confirmed cases of COVID-19 from December 31, 2019, to February 17, 2020. Results: The new-suspected COVID-19 case numbers were significantly correlated with lagged series of SMSI. SIMI could be earlier detected 6–9 days than new-suspected cases of COVID-19. The optimal method was subset select, which had lowest estimation error and a moderate number of predictors. The subset selection method was also significantly correlated with new-confirmed COVID-19 cases after validation. SMSI findings on lag day 10 were significantly correlated with new-confirmed COVID-19 cases. Conclusions: SMSI could be a significant predictor of number of COVID-19 infections. Funding Statement: Lo-Hsu Medical Foundation, Lotung Poh-Ai Hospital, supports SzuYuan Wu’s work (Funding Number: 10908 and 10909). Lei Qin's work is supported by University of International Business and Economics Huiyuan outstanding young scholars research funding (17YQ15), the Fundamental Research Funds for central Universities in UIBE (CXTD10-10). Ben-Chang Shia’s work was also supported by an institutional grant from Taipei Medical University (Taipei, Taiwan) for New Faculty Research (TMU103-AE1-B22). Declaration of Interests: The authors have no potential conflicts of interest to declare.