ABSTRACT
Combatting misinformation is an important part of the global effort to fight against COVID-19. In this paper, we first present a large-scale, publicly available dataset named COVMIS for research on COVID-19 misinformation. COVMIS was constructed to support the misinformation identification approach that mimics the act of fact checking by human for truth labelling. COVMIS is collected from November 2019 to March 2021, this dataset contains 14, 384 claims (statements), 134, 320 related articles, and many features associated with the claims such as claimants, news sources, dates, truth labels (true, partly true or false) and justifications for the truth labels. Each claim is associated with a set of related articles that were collected from reputable sources and serve as the ground truth to assess the validity of the claim. We provide statistics and a detailed analysis of the dataset, and discuss a variety of its potential use cases. Using COVMIS, we then obtained new experimental results illustrating methods that can be used to significantly improve the performance of the fact checking approach for misinformation identification. © 2022 IEEE.
ABSTRACT
Scattering of topologically structured light is highly sensitive to the position of a scattering object. We show that the position of a coronavirus-like 100 nm polystyrene sphere can be measured optically with deeply subwavelength accuracy. © Optica Publishing Group 2022, © 2022 The Author(s)