SC1.35 ECS

Machine learning (ML) is a well-established approach to complex data analysis and modelling in different scientific fields and in many practical applications. Nowadays, ML algorithms are widely used as efficient tools in GI Sciences, remote sensing, environmental monitoring and space-time forecasting. The short course gives an overview of ML algorithms widely applied in data exploration and modelling of high dimensional and multivariate geoscientific data. The main topics of the course, presented within the framework of a generic data-driven methodology of modelling, include detection of patterns and predictability, feature selection, unsupervised, supervised and active learning, visual analytics. Real case studies consider environmental pollution, natural hazards and renewable energy resources assessments.

Share:
Co-organized as ERE8.9/NH10.6/NP10.5
Convener: Mikhail Kanevski | Co-conveners: Vasily Demyanov, Fabian Guignard
Wed, 10 Apr, 14:00–15:45
 
Room -2.31
Machine learning (ML) is a well-established approach to complex data analysis and modelling in different scientific fields and in many practical applications. Nowadays, ML algorithms are widely used as efficient tools in GI Sciences, remote sensing, environmental monitoring and space-time forecasting. The short course gives an overview of ML algorithms widely applied in data exploration and modelling of high dimensional and multivariate geoscientific data. The main topics of the course, presented within the framework of a generic data-driven methodology of modelling, include detection of patterns and predictability, feature selection, unsupervised, supervised and active learning, visual analytics. Real case studies consider environmental pollution, natural hazards and renewable energy resources assessments.