Statistical supports for mining sequential patterns and improving the incremental update process on data streams

Recently the knowledge extraction community takes a closer look to new models where data arrive in timely manner like a fast and continous flow, i.e. data streams. As only a part of the stream can be stored, mining data streams for sequential patterns and updating previously found frequent patterns need to cope with uncertainty. In this paper, we introduce a new statistical approach which biaises the initial support for sequential patterns. This approach holds the advantage to maximiez either the precision or the recall, as chosen by the user, and limit the defradation of the other criterion. Moreover, these statistical supports help building statistical borders which are the relevant sets of frequent patterns to use into an incremental mining process. Theoretical results show that the technique is not far from the optimum, from the statistical standpoint. Experiments performed on sequential patterns demonstrate the interest of this approach and the potential of such techniques.