Table 4.

Thematic clusters based on keyword co-occurrence.

Cluster	Cluster 1 (red)	Cluster 2 (green)	Cluster 3 (blue)
Cluster theme	Disinformation in social media	Techno-scientific research for auto-detection of fake news	COVID-19-induced infodemics
TP	3273	4293	3983
TC/TP	15.6	15.6	17.9
TC	51141	66856	71370
Top ten keywords	fake news disinformation twitter media literacy fact checking post-truth journalism facebook political communication trust	machine learning deep learning natural language processing deep fake artificial intelligence rumour sentiment analysis text classification neural network support vector machine	social media misinformation covid-19 infodemic coronavirus pandemic vaccination internet public health health communication vaccine hesitancy
Top three cited articles, their focus, number of citations	Del Vicario, M.D. et al. Misinformation online TC = 1029 Wang, Y. et al. Health Misinformation on Social Media TC = 589 Zubiaga, A. et al. Rumours in social media TC = 79	Tandoc, E.C. et al. Typology for fake news TC = 981 Rossler, A. et al. Machine learning detection TC = 977 Ruchansky, N. et al. Deep learning for fake news TC = 588	Kata, A. Anti-vaccine TC = 584 Cinelli, M. et al. COVID-10 infodemic TC = 742 Pennycook, G. et al. COVID-19 Misinformation TC = 717

Next, we analyze the top three cited articles from each of the clusters.