Skip to content

A python library to fetch, filter and normalize open data related to COVID and excess mortality

License

Notifications You must be signed in to change notification settings

fraunhoferportugal/fairisk

Repository files navigation

License: CC BY-NC 4.0 PyPI - Python Version PyPI Downloads DOI

FAIRisk | Improving risk estimation with open resources

FAIRisk combines open-source globally available data related with risk scales and country preparedness for epidemic crisis with post-COVID-19 data. It aims to facilitate the creation of preventive insights at country-level and the suggestion of improvements to current risk modelling strategies, and assist the work of the scientific community and decision-makers dealing with COVID-19 (or related) crisis.

This repository addresses data interoperability challenges of fetching and combining several openly available sources of multimodal data, so these can be coherently used from a centralized data model. This model was designed bearing FAIR principles and EU's open data guidelines in mind to promote an adequate, well-documented and simplified use of the combined data.

Approach

A semantic data model was defined, following the analysis of several relevant sources, in order to typify and indentify the most relevant concepts, while attempting to maximize its generalization. Data from 6 different sources was organized and merged in a single data model, where each country entity is represented by up to 6 categories:

  • Demographics: population by age group.
  • Indicators: raw indicators' data measured for each country.
  • Scores: indexes, scales, or scores that assist the comparison of qualitative or estimated parameters across countries.
  • Mortality: count of deaths by age group.
  • COVID-19: number of cases, deaths, tests, ICU patients, hospitalizations, vaccinations, and stringency index due to the COVID-19 pandemic.
  • Mobility: statistics of citizens movement estimations.

Check out our architecture documentation for more details.

Getting started

You can install our python package using pip:

pip install fairiskdata

We have compiled a sample notebook to demonstrate the use of the library and its methods.

For additional information check the getting started guide or the full documentation.

License

All visualizations and code available in this repository are licensed under the Creative Commons BY-NC-SA 4.0 license.

All data fetched by the methods available in this repository was produced by third-parties and is subject to the license terms from the original third-party authors. The sources from which data was fetched are kept and made available as metadata at all stages. Sources are also detailed here. You should always check the license of all third-party data before use.

Funding

The authors would like to acknowledge the financial support obtained from EOSCsecretariat.eu. EOSCsecretariat.eu has received funding from the European Union's Horizon Programme call H2020-INFRAEOSC-05-2018-2019, grant Agreement number 831644.

EOSC logo

Authors

FhP logo

These resources were developed by Fraunhofer AICOS.

Development team: Diana Gomes ([email protected]), Catarina Pires ([email protected]), David Ribeiro ([email protected]), Duarte Folgado ([email protected]), Ricardo Santos ([email protected]), Telmo Barbosa ([email protected]).

About

A python library to fetch, filter and normalize open data related to COVID and excess mortality

Resources

License

Stars

Watchers

Forks

Packages

No packages published