The project concerns the development of data science methods and their application in the domain of fusion energy. The Research unit Nuclear Fusion (infusion) has been at the forefront of this research field since 20 years. The present activities concern pattern recognition in fusion databases, probabilistic modeling of stochastic plasma phenomena, joint processing of data from multiple sensors (sensor fusion) and detection of unexpected events and failure of machine components for optimizing maintenance strategies (anomaly detection and predictive maintenance).