LASIGE researchers, Márcia Barros, Diana Sousa, Pedro Ruas, Francisco M. Couto, and the former member André Lamúrias, presented their work “COVID-19: A Semantic-Based Pipeline for Recommending Biomedical Entities” at the workshop (part 2) at the “2020 Conference on Empirical Methods in Natural Language Processing” (EMNLP2020), held virtually, 16-20 November 2020.

LASIGE’s researchers mobilized themselves and joined the global research community to respond with timely research and scientific analysis that can contribute to our understanding and management of the coronavirus (COVID-19). This work aims to build a new semantic-based pipeline for recommending biomedical entities to scientific researchers.

With the increasing number of publications about COVID-19, it is a challenge to extract personalized knowledge suitable for each researcher and valuable insights and information may be contained in vast quantities of text and speech data. The main contributions of this work are a dataset of 9k articles automatically annotated with relevant items/concepts for CORD19; a sample dataset curated for CORD-19; a sample dataset with relations between the entities of the four ontologies; an implicit feedback matrix based on the previous datasets.

The paper is available: here.