LIST takes up the challenge of EGC 2018

As part of its environmental research activities and as a member of the International French-speaking Extraction and Knowledge Management Association (Association EGC), the Luxembourg Institute of Science and Technology (LIST) participated in the EGC 2018 conference, held in Paris (FR) from 22 to 26 January.

The Extraction and Knowledge Management (EGC) conference is an annual event bringing together researchers and practitioners in disciplines related to data and knowledge science. The conference, which took place for the 18th time, counted around 200 participants, and provided the opportunity to discuss with colleagues from organizations such as the INRIA, the CNRS, etc.

LIST was represented by three members of the Environmental Research and Innovation department (ERIN): Pierrick Bruneau, Philippe Pinheiro and Yoann Didry, specialists in data analysis, who participated in the EGC 2018 challenge, a challenge in the sun on the island of Reunion. During the conference, their article, entitled “Prédiction du Rayonnement Solaire par Apprentissage Automatique” (Solar Radiation Forecast with Machine Learning), was selected and won the Challenge Award.

The LE2P laboratory (Energetics, Electronics and Processes) and the Computer and Mathematics Laboratory of the University (LIM) of La Reunion offered to analyse the flow/radiation data on the island of Reunion. This project is part of a development policy towards energy autonomy on the island of Reunion. A history of multi-source sensors over several years was made available to researchers in the form of multivariate time series. The aim was to address the challenge of exploratory or predictive analyses of these data. The LIST experts therefore applied data analytics methods to the data and described their research results in the communication selected.

This work is in line with the research activities in e-Science at LIST. It aims to highlight the application of methods relevant to statistical learning to spatio-temporal data, as this type of data is common in the ERIN department (for example, hydrological and meteorological data). These kinds of developments could also benefit internal synergies with the spatio-temporal data management platform iGuess, developed by LIST.

Congratulations to LIST researchers for taking up this challenge!