A Corpus for Entity Recognition in COVID-19 Full-text Literature

ABSTRACT

To support the development of entity recognition tools, this study manually annotates 99 full-text articles about COVID-19. Each article is annotated by 6 annotators through two rounds. 18 types of entity are involved, including genes, diseases, chemicals, coronaviruses and so on. We also calculate the inter-annotator agreement (IAA) scores in term of multi-κ measure to ensure the quality of the annotations. In the end, 39, 118 entity mentions are manually annotated in total. © Copyright 2022 for this paper by its authors.