Nikos Salamanos, Pantelitsa Leonidou, Nikolaos Laoutaris, Michael Sirivianos, Maria Aspri, Marius Paraschiv

In light of the growing impact of disinformation on social, economic, and political landscapes, accurate and efficient identification methods are increasingly critical. This paper introduces HyperGraphDis, a novel approach for detecting disinformation on Twitter that employs a hypergraph-based representation to capture (i) the intricate social structure arising from retweet cascades, (ii) relational features among users, and (iii) semantic and topical nuances. Evaluated on four Twitter datasets -- focusing on the 2016 U.S. Presidential election and the COVID-19 pandemic -- HyperGraphDis outperforms existing methods in both accuracy and computational efficiency, underscoring its effectiveness and scalability for tackling the challenges posed by disinformation dissemination. The HyperGraphDis displayed exceptional performance in an evaluation using a COVID-19-related dataset, achieving an impressive F1 score of approximately 92.5%. This result represents a notable improvement of around 7% compared to other existing methods. Additionally, significant enhancements in computation time were observed for both model training and inference processes. In terms of model training, completion times were noticeably accelerated, ranging from 1.6 to 16.5 times faster than previous benchmarks. Similarly, during inference, computational times demonstrated increased efficiency, ranging from 1.3 to 17.7 times faster than alternative methods.