Section: New Results
NTGSP: Mining Negative Temporal Patterns
Participants: K. Tsesmeli, M. Boumghar, T. Guyet, R. Quiniou, L. Pierre
In  the authors study the problem of extracting frequent patterns containing positive events, negative events specifying the absence of events as well as temporal information on the delay between these events.  defines the semantics of such patterns and proposes the NTGSP method based on state-of-the-art approaches. The performance of the method is evaluated on commercial data provided by EDF (Électricité de France).