Aggressive behaviour of anti-vaxxers and their toxic replies in English and Japanese | Humanities and Social Sciences Communications
https://www.nature.com/articles/s41599-022-01245-x
This paper analyzed English and Japanese Twitter data to characterize the reply behavior of anti-vaccine (anti-vaccine) groups. The main findings are as follows
- Anti-vaccinationists are the most aggressive in sending replies to users of other beliefs, especially neutral accounts, which are significantly more toxic and negative in content. These replies are often directed to vaccine-related posts.
- Influential users and accounts with large followings (e.g., medical professionals, scientists, policy makers, and media representatives) tend to receive the most toxic replies from anti-vaccine groups.
- Pro-vaccine accounts with only a few followers in English also received highly toxic replies, but this was not the case in Japanese.
These results provide insight into language-dependent and language-independent measures to counter the aggressive behavior of antivaccinationists. They suggest that platforms should take steps to prevent exposure of antivaccinationist discourse to other users and that pro-vaccinationists should be prepared for highly toxic and emotional replies from antivaccinationists.
What are the specific methods?
The following specific methods were used in this study
- Data collection: continuous collection of COVID-19 related tweets from February to December 2020. Extracted tweets containing vaccine-related keywords.
- Clustering based on RT network: community detection was performed on the RT network to classify users into stances against vaccines (anti-vaccine, neutral, etc.).
- Analyze reply behavior: examine the frequency of replies from anti-vaccine groups, target clusters, popularity of reply destinations, etc.
- Toxicity measurement: Google’s Perspective API was used to measure the aggressiveness of replies. Japanese tweets were measured for toxicity after translation.
- Emotional analysis: the degree of negativity and positivity of replies and emotional arousal were measured using the Emotional Dictionary (LIWC 2015, Wariner et al.
- Analysis of the relationship between the number of followers and the toxicity: the correlation between the number of followers of the user to whom the reply was sent and the maximum toxicity value of the reply received was investigated.
By utilizing these methods, we analyzed the characteristics of aggressive reply behavior of anti-vaccine groups from multiple perspectives in the different language spheres of English and Japanese.
This page is auto-translated from [/nishio/Aggressive behaviour of anti-vaxxers and their toxic replies in English and Japanese](https://scrapbox.io/nishio/Aggressive behaviour of anti-vaxxers and their toxic replies in English and Japanese) using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.