Quality-Aware Mining of Data Streams.
Conny Franke, Michael Hartung, Marcel Karnstedt, Kai-Uwe Sattler · ICIQ · 2005
Due to the inherent characteristics of data streams, appropriate mining techniques heavily rely on window-based processing and/or (approximating) data summaries. Because resources such as memory and CPU time for maintaining such summaries are usually limited, the quality of the mining results is affected in different ways. Based on selected mining techniques, we discuss in this paper relevant quality measures for analysis results. Furthermore, we describe extensions to two specific stream mining algorithms allowing (1) to estimate resource consumptions (mainly memory space) based on user-specified quality requirements and (2) to determine the output quality for changes in the available resources.