Disinformation of text mining online about tobacco and the COVID-19 discussed on Sina Weibo

INTRODUCTION: During the COVID-19 pandemic, various types of disinformation have emerged from the media. This study focuses on the online disinformation about tobacco and the COVID-19 on the Sina Weibo, the Chinese largest new media microblog platform. METHODS: The related posts from the beginning of the epidemic in December 2019 to 19 January 2021 were searched. Text mining technology was applied on these posts to identify content on ‘smoking can prevent COVID-19’. Descriptive research was used to analyze the dataset. RESULTS: Among the 912 original posts, 508 informative posts were selected after artificial recognition, including 112 posts of spreading disinformation and 396 which dispel the disinformation. Of the disinformation posts, 74% (83/112) cited the results of scientific research, and 17% (19/112) mentioned that smog from burning Asian wormwood could prevent COVID-19. By analyzing the public’s comments on these 112 disinformation posts, it was suggested that about 12% of the comments were in support, and 88% of the posts were opposed or invalid. The proportion of supportive comments on pseudo-scientific information was higher than on plain disinformation, 21% and 9%, respectively. CONCLUSIONS: The disinformation of promoting smoking as a way to prevent COVID-19 has the typical feature of using pseudo-scientific arguments to package disinformation, making it very difficult for readers without professional knowledge to identify. Such actions harm both tobacco control and COVID-19 prevention.