Potential limitations in COVID-19 machine learning due to data source variability: A case study in the nCov2019 dataset | Zendy

Carlos Sáez | Zendy; Nekane Romero-García | Zendy; J. Alberto Conejero | Zendy; Juan M. GarcíaGómez | Zendy

AI Assistant Blog Pricing

Home ZAIA Blog

Open Access

Potential limitations in COVID-19 machine learning due to data source variability: A case study in the nCov2019 dataset

Author(s) -

Carlos Sáez,

Nekane Romero-García,

J. Alberto Conejero,

Juan M. GarcíaGómez

Publication year - 2020

Publication title -

journal of the american medical informatics association

Language(s) - English

Resource type - Journals

SCImago Journal Rank - 1.614

H-Index - 150

eISSN - 1527-974X

pISSN - 1067-5027

DOI - 10.1093/jamia/ocaa258

Subject(s) - overfitting , computer science , data quality , machine learning , covid-19 , representativeness heuristic , artificial intelligence , bottleneck , data sharing , data source , data science , data mining , disease , medicine , artificial neural network , statistics , infectious disease (medical specialty) , pathology , metric (unit) , operations management , alternative medicine , mathematics , economics , embedded system

The lack of representative coronavirus disease 2019 (COVID-19) data is a bottleneck for reliable and generalizable machine learning. Data sharing is insufficient without data quality, in which source variability plays an important role. We showcase and discuss potential biases from data source variability for COVID-19 machine learning.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.

Having issues? You can contact us here

Accelerating Research