TRUE

TRUE aims to improve the way changing data streams are interpreted, for example the ones produced by industrial systems, networks, or other cyber-physical environments. The idea is to understand not only what is happening, but also which causes are most plausible and which risks may follow.

To do that, the project combines models that learn from data, methods for comparing different causal explanations, and tools for estimating how reliable a prediction really is. In simple terms, it tries to make real-time decision support both easier to understand and more cautious.