In a recent study published in PLOS ONE, researchers analyzed the misinformation of coronavirus disease 2019 (COVID-19) on Twitter.
Study: analysis of the misinformation of COVID-19 on Twitter using the hashtags #scamdemic and #plandemic: retrospective study. Image credit: rafapress / Shutterstock
Fund
Widespread use of social media during the COVID-19 pandemic had led to a “infodemic” of misinformation and misinformation about COVID-19, leading to potentially fatal consequences. Understanding the magnitude and impact of this false information is essential for public health agencies to estimate the behavior of the general population with respect to vaccine adoption and non-pharmaceutical interventions (NPI) such as social distancing and ‘masking.
About the study
In the present study, the researchers evaluated the tweets circulating on Twitter containing the hashtags #Plandemic and #Scamdemic.
On January 3, 2021, the team used Twint, a Twitter tracking tool, to collect tweets in English containing the hashtags #Plandemic or #Scamdemic posted between January 1 and December 31, 2020. On January 15, 2021, the team subsequently used the Twitter Scheduling Software (API) app to get the same tweets using the corresponding tweet identities. The team provided descriptive statistics for selected tweets, such as correlated tweet content and user profiles, to determine the availability of tweets in both datasets developed according to Twitter API status codes .
The sentiment analysis of the tweets was done by tokenizing the tweets and cleaning them up. Later, tokens were transformed into their root form using natural language processing techniques, such as lemmatization, derivation, and the elimination of limited words. The Python VADER library was used to recognize and classify the feeling of the tweet as neutral, positive, or negative, and the subjectivity of the tweet as subjective or objective. VADER applied a rules-based analysis of feelings with a polarity scale ranging from -1 to 1.
Subjective analysis was performed using TextBlob, which tagged each tweet on a scale of zero or objective to one or subjective. Targeted tweets were considered to provide facts, while subjective tweets communicated an opinion or belief. The team viewed a histogram of the subjectivity scores of the hashtags #Plandemic and #Scamdemic. The Python library was also used to label the main emotion associated with each tweet as fear, anticipation, anger, surprise, confidence, sadness, joy, disgust, positive or negative.
The predominant topics covered in the tweet library were recognized and an machine learning algorithm was applied. This algorithm identified the tweet groups using a representative group of words. The words with the highest weight in each group were used to define the content of each topic.
Results
The results of the study showed that a total of 420,107 tweets included the hashtags #Plandemic and #Scamdemic. The team removed tweets that were retweets, replies, non-English or duplicates to retain 227,067 tweets from approximately 40,081 users. Almost 74.4% of total tweets were posted by 78.4% of active Twitter users, while 25.6% of tweets were posted by 21.6% of users whose account was suspended on January 15, 2021. The team noted that probably users with suspended profiles. to tweet more. Users who used both hashtags had a 29.2% chance of being suspended, compared to 25.9% of tweets using #Plandemic and 13.2% of tweets using #Scamdemic.
The team found that most users were 40 or older. In addition, suspended users mostly include men and users 18 years of age or younger and 30 to 39 years old. Almost 88% of active users and 79% of suspended users tweeted from their personal accounts. In particular, almost 65% of the tweets analyzed showed objectivity.
Analysis of the emotions in the tweets revealed that fear was the predominant emotion, followed by sadness, confidence, and anger. Emotions such as surprise, disgust, and joy were the least expressed, while suspended tweets were more likely to show disgust, surprise, and anger.
The general sentiment expressed by the tweets containing the hashtags #Plandemic and #Scamdemic was negative. Overall average weekly feelings were -0.05 for #Plandemic and -0.09 for #Scamdemic, where 1 and -1 denoted completely positive and negative feelings, respectively.
The most frequently observed tweet topic was “complaints against warrants introduced during the COVID-19 pandemic,” which also included complaints against masks, closures, and social distancing. They then posted tweets with topics “ministering to the dangers of COVID-19,” “lies and brainwashing by politicians and the media,” and “corporations and the global agenda.”
Overall, the results of the study showed that COVID-19-related tweets showed an overall negative sentiment. Although several tweets expressed anger against the restrictions during the pandemic, a significant proportion of tweets also presented misinformation.