There have been increasing concerns about spreading COVID-19 fake news and misinformation from social media sites (SNSs) such as Facebook and Twitter, as they facilitate connection and communication on a large scale. Because of the massive amount of information transmitted through SNSs, manual verification of such information is impossible, prompting the development and implementation of automated methods for fake news identification, aka automatic fact-checking. Fake news creators employ a variety of aesthetic tactics to increase their success rates, one of which is to excite the readers' sentiment. Therefore, this research uses sentiment analysis to analyze whether sentimental and emotional words in SNSs content could explain the situations between the spreading of true and fake news. In this way, governments and platform providers could take action to help the general public identify fake news and misinformation and curb them at their source. This research also offers insights to the public on the importance and impacts of sentiment words in SNS content.
TopIntroduction
Nowadays, people receive information through various means. Particularly, young people rely on social and electronic media, instead of traditional media, to know what is happening worldwide (Yu et al., 2021; Ding et al., 2021; Au et al., 2022). This is further facilitated by anytime, anywhere, and ubiquitous mobile Internet access (Lam et al., 2022; Chan et al., 2022). As traditional media usually have the freedom to monitor their content and is edited by professional journalists to ensure professional and ethical standards, traditional media also serve as gatekeepers to provide accurate information to the public (McElroy, 2013; Skovsgaard & van Dalen, 2013). Nevertheless, traditional media are no longer the primary source of information, especially for the younger generation.
Today, we spend a large portion of our time on the Internet, even anywhere, anytime with mobile devices, through social networking sites (SNS), ranging from working and socializing to entertaining (Wang et al., 2021; Dong et al., 2021; Wong et al., 2022). As a result, we are exposed mainly to online information rather than physical ones (Yu et al., 2021; Wang et al., 2016). Despite the popularity of SNS, information accuracy is not guaranteed (Ho et al., 2022). Yet, there are more serious SNS applications, such as communities of practice (Lei et al., 2021) and academic SNS (Yang et al., 2022). Before release, such information does not have a review mechanism and requires SNS to formulate policies on monitoring inappropriate information, including fake news, misinformation, and hateful speech. They often rely on user reports for suspected content investigation after posting, which may be too late to stop spreading fake news and misinformation. Therefore, the public needs tools, like fact-checking mechanisms, for identifying the characteristics of fake news and misinformation fact-checking (Au et al., 2021a; 2021b; 2021c; Guo et al., 2022).
This paper uses ‘misinformation’ and ‘fake news’ interchangeably. Fake news is “the deliberate presentation of false or misleading claims as news, where the claims are misleading by design” (Gelfert, 2018, p. 108), while misinformation is released by mistake without bad intention (Ho et al., 2022; Au et al., 2021a). It aims to create a social, political, and economic bias in people’s minds for others’ benefit, affecting and exploiting people by making misleading information that sounds authorized (Shu et al.,2017). It may lead to a public health crisis or even a riot on the extreme (Ho et al., 2022; Au et al., 2021a; 2021c). The unexpected COVID-19 pandemic has caused an outburst of misinformation about the disease, and SNSs are essential channels for receiving, sharing, and posting information regarding the pandemic (Naeem et al., 2020; Ho et al., 2022).
There are many examples of fake news and misinformation about healthcare issues. Examples like fake cures of gargling with salt water and injecting with bleach and 5G cellular network can cause the pandemic can be easily found online (Bavel et al.,2020). Another example is former US President Trump embracing hydroxychloroquine, an anti-malarial drug against the new COVID-19, with unproven medical evidence on Twitter (Paz, 2020). Besides, specific words in headlines like “die,” “death,” “disaster,” and “end of the world” may arouse the spreading of news as those words are life-related and induce fear in the readers (Paz, 2020). Those topics and words may lead to fake news spreading on a broader scope.