Clustering Multivariate Climate Data Streamsusing Fractal Dimension

A data stream is a flow of data produced continuously along the time. Storing and analyzing such information become challenging due to exponential growth of the data volume collected. In this context, some methods were proposed to cluster data streams with similar behavior along the time. However, those methods have failed on clustering data flows with more than one attribute, i.e., multivariate flows. This paper introduces a new method to cluster multivariate data streams, based on fractal dimension, reading the data only once. We evaluated our method over real multivariate data streams generated by climate sensors. Not only was our method able to cluster the flows of data, but also identified sensors with similar behavior during the analyzed period.