2006 · 56 citations · 5 references
ComputingData ScienceData AnalyticsData MiningTime Series AnalysisCluster DetectionTemporal Data MiningTime Series ClusteringClusteringConcept DriftData Stream MiningDivisive ClusteringHierarchical ClusteringHierarchical Time Series ClusteringHierarchyOnline Time Series Clustering
This paper presents a time series whole clustering system that incrementally constructs a tree-like hierarchy of clusters, using a top-down strategy. The Online Divisive-Agglomerative Clustering (ODAC) system uses a correlation-based dissimilarity measure between time series over a data stream and possesses an agglomerative phase to enhance a dynamic behavior capable of concept drift detection. Main features include splitting and agglomerative criteria based on the diameters of existing clusters and supported by a significance level. At each new example, only the leaves are updated, reducing computation of unneeded dissimilarities and speeding up the process every time the structure grows. Experimental results on artificial and real data suggest competitive performance on clustering time series and show that the system is equivalent to a batch divisive clustering on stationary time series, being also capable of dealing with concept drift. With this work, we assure the possibility and importance of hierarchical incremental time series whole clustering in the data stream paradigm, presenting a valuable and usable option.
5
Data mining: concepts and techniques
Jiawei Han, Micheline Kamber · Choice Reviews Online · 2012 · 28.9K citations
Mining high-speed data streams
Pedro Domingos, Geoff Hulten · 2000 · 2.2K citations · Full text
Learning in the Presence of Concept Drift and Hidden Contexts
Gerhard Widmer, Miroslav Kubát · Machine Learning · 1996 · 1.4K citations · Full text
Requirements for clustering data streams
Daniel Barbará · ACM SIGKDD Explorations Newsletter · 2002 · 153 citations