Fabian Guignard

Research Topic

My current research interests are related to large space-time datasets and data science methodologies, including:

  • machine learning and data mining,
  • spatial and spatio-temporal statistics,
  • methodological developments for the quantification of spatio-temporal prediction uncertainties,
  • nonlinear time series analysis,
  • practical applications on real-world large datasets,
  • development of Python modules.

Applications of these topics among others are environmental science and renewable energy. In fact, I am involved in a project supported by the National Research Programme 75 “Big Data” (NRP 75) of the Swiss National Science Foundation (SNSF). This project, in collaboration with EPFL, aims to estimate the hybrid renewable energy potential of urban areas, including forecasting and uncertainty quantification.

Biography

After seven years practising geomatics in environmental engineer firms, I was trained as a mathematician at the Swiss Federal Institute of Technology in Lausanne (EPFL). During my studies, I was strongly interested in quantitative disciplines (statistics, ranking and sorting algorithms, convex optimisation), but curious about bridges between mathematics and environment (cartography, geodesy). In 2017, I joined the GeoKDD group as a doctoral researcher in data science.