Abstract
Mining temporal knowledge has many applications. Such knowledge can be all the more interesting as some time constraints between events can be pushed into during theminingtask. As well in data mining as in machine learning, some methods have been proposedto extract and manage such knowledge using temporal constraints. In particular some work has been done to mine generalized sequential patterns. However such constraints are often too crisp or need a very precise assessment to avoid erroneous information. Within this context, we propose an approach based on graphs of sequences derived from extended temporal constraints. These relaxed constraints enable us to find more generalized sequential patterns. We also propose a measure of the temporal accuracy of the extracted sequences compared to the initial constraints; this measure will provide the user with a tool to analyse the numerous extracted patterns.