Machine learning for COVID-19 literature
Machine learning and text mining techniques to assist classification of literature for our COVID-19 repository.
Read moreAbout us
Motivated by the current pandemia, a group of academics got together to put their efforts in organizing COVID collections of papers and process them to facilitate their access. Mónica Ballesteros, who at the beginning of this project was working for the Vall d'Hebron Hospital in Barcelona, together with Johana Caro* from the Agency for Quality and Health Assessment of Catalonia (AQuAS), Juliana Sanabria*, and Johannes Graën from Pompeu Fabra University and Gothenburg University built a website in internet offering COVID-19 literature pre-organized in different medical categories and subcategories with the aim of facilitating professionals to quickly find what has been published within their domain of specialization.
One of the initial purposes was to make the literature more easily accessible for clinicians, in particular in Spanish speaking countries. Therefore translations into Spanish of the abstracts were provided, using a combination of manual and semi-automated approaches. The translations are made by volunteer collaborators under the direction of Antoni Oliver González from the Open University of Catalonia (UOC).
Later on the team expanded to include researchers with expertise in natural language processing of biomedical literature. The collaboration includes people of the Program of Computational Genomics of the Center for Genomics Sciences, UNAM, Mexico. Specifically, Oscar Lithgow, Carlos-Francisco Méndez-Cruz, and Julio Collado-Vides; Alejandra López, PhD student in the Terminology Program of the Pompeu Fabra University in Barcelona; Fabio Rinaldi, NLP expert from the Swiss AI lab IDSIA and group leader at the Swiss Institute of Bioinformatics and Oscar Lithgow who is also part of the IDSIA.
*At the present time, Johana Caro and Juliana Sanabria are no longer contributors of this project.