A Clustream strategy for Functional Boxplots on multiple streaming time series

"In this paper we propose a micro-clustering strategy for Functional Box-. plot variables defined on multiple streaming time series splitted in non overlapping. windows. It is a two step strategy. In the first step it performs an on-line summa-. rization keeping updated the set of functional data structures, named Functional. Boxplot micro-clusters; in the second step it reveals the final summarization by pro-. cessing, off-line, the Functional Boxplot micro-clusters. Thus a new definition of. micro-cluster based on using the Functional Boxplot as the centroid is proposed.. Moreover a proximity measure which allows to allocate the data to the new defined. micro-clusters is defined. This will allow to get a graphical summarization of the. streaming time series by five functional basic statistics. The obtained synthesis will. be able to keep track of the dynamic evolution of the streams."