摘要：Knowledge Discovery in Databases (KDD) techniquespresent limitations when the volume ofdata to process is very large. Any KDD algorithmneeds to do several iterations on the complete setof data in order to carry out its work. For continuousdata stream processing it is necessary tostore part of it in a temporal window.In this paper, we present a technique that usesthe size of the temporal window in a dynamic way,based on the frequency of the data arrival andthe response time of the KDD task. The obtainedresults show that this technique reaches a greatsize window where each example of the stream isused in more than one iteration of the KDD task
