Diagnosing, REcommending Actions and Modelling
The main research topics of the Dream team are about aiding monitoring
and diagnosis of time evolving systems. The main issue is to infer the
state of a system from observations provided by sensors in order to
detect and characterize potential anomalies or failures within the
system. We use a model-based approach relying on normal and faulty
behavioral models. These models are temporal qualitative
discrete-event models such as temporal communicating automata,
temporal causal graphs or sets of chronicles.
The aim of the DREAM research team is to design monitoring systems and support decision systems for managing complex entities evolving during time. We rely upon qualitative models and background knowldge, to enable the user/stakeholder to understand more easily the outputs (predictions, diagnoses, recommendations) proposed to him. We are working particularly on the automatic knowledge acquisition (with machine learning and data mining techniques) from massive datasets, with spatial and temporal dimensions. Our aim concerns both designing models of observed systems and recommending actions that could improve the behavior of those systems. Our favorite application domain is environment protection (agro- and eco-systems). We have tight collaboration with INRA (National Initute for Research in Agronomy) and Agrocampus grand Ouest (school of engineering in Agronomy).
See also the presentation (in French) on the INRIA website.