10 - Machine learning for future intelligent air quality networks
- Event
- Sixth Scientific Meeting EuNetAir
2016-10-05 - 2016-10-07
Academy of Sciences, Prague, Czech Republic - Band
- Sixth Scientific Meeting EuNetAir
- Chapter
- Proceedings
- Author(s)
- S. De Vito, E. Esposito, M. Salvato, G. Fattoruso, G. Di Francia - ENEA - Agenzia per le Nuove Tecnologie, l’ Energia e lo Sviluppo Economico Sostenibile, Portici (NA) - Italy
- Pages
- 38 - 41
- DOI
- 10.5162/6EuNetAir2016/10
- Price
- free
Abstract
During the last few years, machine learning emerged as a very effective tool for data analysis and sematic value extraction from the large amount of data generated from deployed chemical multisensors devices. Many works have now highlighted the potential impact on multisensor device calibration, drift counteraction, data assimilation, optimal deployment of these classes of algorithms. Unlike 5 years ago, the huge amount of available data make possible to confirm this potential on real-world long-term deployments. This work analyze the literature produced by EuNetAir partners extracting the lessons cooperatively learnt about their impact and propose a novel architecture for future intelligent air quality networks based on the machine learning emerging paradigm.