Improving data integration
A_Gael Varoquaux_Inria-0304-682_300_400
© Inria / Photo G. Scagnelli

Project: Learning data integration, from discrete entities to signals (LearnI)

Project leader: Gaël Varoquaux – Parietal team – Inria Saclay - Île-de-France

Activity scope: medical imaging and epidemiological studies

Aim: to reduce the amount of time spent by human experts standardising input data for statistical analysis and learning models by hand (generally speaking, the information gathered on patients by different institutions: profession, description and name of illness, socio-economic status, etc.). This involves imagining models for optimising the gathering of multimodal data, enabling statistical analysis tools to operate, even with non-standardised databases.