This study explores the delayed and combinatory responses of online public opinion to the intensity of the COVID-19 pandemic using natural language processing, statistical analysis, and machine learning-based causal inference. Findings indicate a long-term lag in online public opinion's response, with identical COVID-19 intensity data triggering multiple delayed responses and a single delayed opinion datum influenced by multiple preceding data points. This results in a waveform structure of real-world impacts influenced by online public opinion, with sensitivity varying across different time periods.

Yamin Du, Huanhuan Cheng, Qing Liu, Song Tan