Online medical treatment refers to the use of the Internet to provide users with health information services (Deng, &Hu, 2019). In China ,45 million users of online medical treatment in 2019, with a penetration rate of 6.6%. Affected by the 2020 COVID-19 virus, it is estimated that the penetration rate of them will reach 7.9% in 2020 (iimedia, 2020). The main advantages and attractions for it have been found to include access, anonymity, potential for interactivity, and social support (Cline, & Haynes, 2001). However, the Internet is a technology-based medium, and there is no moral obligation involved, people are less likely to trust in the Internet as a source of health information (Ye, 2010). Most studies in the field of social network have focused on people share health information online (Newman, Lauterbach, Munson, Resnick, & Morris, 2011) or Patient–doctor interactions (Bosslet, Torke, Hickman, Terry, & Helft, 2011). At the same time, users will show their attitudes towards online medical treatment to express their emotions. Traditional audience researches are generally done by questionnaire surveys or experiments, but in the digital era, more representative conclusions can be inferred from big data analysis, which can collect and analysis mass data rapidly, especially when public opinion breaks out. This study wants to use a new model on people's attitude by Computer aided information analysis to explores how to use the online public opinion big data to infer the attitude toward online medical treatment of Chinese web-users, and whether there are differences in types of users.
Opinion leaders are essential for users' intentions, health campaigns are no exception (Rogers, & Kincaid, 1981). They usually play an important role on previous studies of healthcare topics (Valente, 2010; Han, & Wang, 2015). Sina Weibo has the largest social media user in China. In Sina Weibo, some people with social awareness, such as scholars, actors, writers, etc. will be shown as “authenticated users” (V-users). On December 7, 2019, the Chinese official media People's Daily post two tweets on Weibo that reflected false information about the cheat of online medical treatments.
By analyzing the all data from the simple sentiment analysis (by snownlp in python) and the LDA classification (by LDAvis in python), it is judged the emotion for distinguishing rate of negative words (<0.3), neutral words (0.3-0.8) to positive words (>0.8) through comparing with the emotion corpus. It found that 43.3% of all users ’attitudes towards online medical care were negative, and the four themes formed at the same time were all negative. After distinguishing between different types of users, emotion negative rate of V users' comments (35.2%) is nearly 10% lower than non-V users’ (44.9%). It means that influencers in social networks send comments tend to use more neutral words than general users. Professional healthcare websites (eg: www.Dingxiangyuan.com) were found receive better reviews than search sites (eg: www.Baidu.com). We also discuss the implications of applying public opinion big data in health information research.