Screenshot
GIF
COVID-19 Papers

CodeVsCovid19

Repository for the hackathon CodeVsCovid19

This project aims to provide a visual way to browse through the papers related to the different coronaviruses. We developped a graph structure of a subset of these papers to show how this can be an interesting visualization for researchers that want to quickly have an overview of the existing papers.

We based our project on https://github.com/gsarti/covid-papers-browser and we use some of their code (located inside of the scripts folder), but we actually went in a totally different direction and we mostly worked on the graph-based visualization. We also created a UI for the search engine: https://onefork.github.io/covid-19/

In the future we aim to provide this tool for the whole dataset.

The visualization is available at: https://covid-19-papers-react.s3.us-east-2.amazonaws.com/index.html

Built With

javascript
jupyter-notebook
python

Submitted to

#CodeVsCovid19

Created by

I coded the front end.

Bin Zhang
MD and Engineer by training, curious HACKER at heart. Reach me via LinkedIn
I worked on the NLP part of the project. I inferred the language of the papers based on their abstract. I also performed topics extraction on the dataset in order to cluster the papers.

Tina Rey
I worked on improving and cleaning the CORD-19 dataset. I also started improving the covid paper browser. Finally I used my skills in deep learning to guide the research in NLP networks

Nicolas Marchal
I worked on the graph construction, embeddings of the abstracts of the papers, I investigated graph-based clustering and spacial clustering and started basic NLP that did not make it to this version of the project.

Valérian Rey