- From: Franck Michel <franck.michel@inria.fr>
- Date: Mon, 4 May 2020 11:13:03 +0200
- To: public-web-vs-covid19@w3.org, Fabien Gandon <fabien.gandon@inria.fr>
- Message-ID: <48f99f4a-3896-433d-900c-2244313fca95@inria.fr>
Dear "Web fights Covid19" CG members,
We would like to let you know about the "Covid On The Web" project,
launched by the Wimmics team (University Cote d'Azur, Inria, CNRS), that
aims to adapt and combine several tools to make it easier to access and
make sense of data related to the Covid-19.
On the one hand, the ACTA platform (Argumentative Clinical Trial
Analysis) is designed to help clinicians analyze clinical trials and
make decisions by automating the extraction of argumentative components
and their visualization as graphs. On the other hand, the CORESE and
MGExplorer platforms allow to deal with knowledge graphs and visualize
them on the web.
"Covid On The Web" proposes to adapt these tools to a continuously
enriched corpus and knowledge graph related to the corona family of
viruses. The project leverages Wimmics' skills and scientific expertise
on knowledge representation, information extraction from text, data and
argument mining, data querying and visualization.
To start with, we have published the "CORD-19 Named Entities Knowledge
Graph" (CORD19-NEKG), an RDF dataset that we have built using the
Entity-fishing et DBpedia Spotlight tools to identify and disambiguate
named entities mentioned in 45000+ scientific articles. Currently, this
amounts to 2.6 million named entities described in RDF and linked to
DBpedia and Wikidata (and hence Wikipedia and beyond). The dataset is
described at and downloadable from
https://github.com/Wimmics/cord19-nekg, and it can also be queried
directly from our Virtuoso SPARQL endpoint:
https://covid19.i3s.unice.fr/sparql. We registered this dataset in the
classification tool Google Doc
<https://docs.google.com/spreadsheets/d/1lWW9sjNBpMqT1k2C83jm6kHWtIPF3jkGNlWCnGgx2ZA/edit#gid=0>
previously announced
<https://lists.w3.org/Archives/Public/public-web-vs-covid19/2020Apr/0009.html>
by Ignacio.
Please do not hesitate to let us know your thoughts and feed-back on
this project.
Regards,
Franck, for the Wimmics team.
Received on Monday, 4 May 2020 09:39:03 UTC