An AI-powered unsupervised error detection framework for large data flows

SOON (Station Observation Outlier fiNder) addresses the challenge of automatically detecting erroneous station measurements in a real-time data flow. To do so, it makes use of a combination of machine learning algorithms belonging to the family of “anomaly detection” which individually leverage on specific information domains: spatial, temporal and parameters consistency domains. This approach allows to efficiently combine these different sources of information and to make the process resilient to the eventual absence of one of them.

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 779790

Client: European Data Incubator
Year: 2018-2019
Location: Europe
Team: Sara Dal Gesso, Marcello Petitta, Marco Venturini, Marco Cucchi, Elisa Arnone