Clustering Multidimensional Sequences in Spatial and Tempor...

Clustering Multidimensional Sequences in Spatial and Temporal Databases

article
Ira Assent, Ralph Krieger, Boris Glavic, Thomas Seidl
Many environmental, scientific, technical or medical database applications require effective and efficient mining of time series, sequences or trajectories of measurements taken at different time points and positions forming large temporal or spatial databases. Particularly the analysis of concurrent and multidimensional sequences poses new challenges in finding clusters of arbitrary length and varying number of attributes. We present a novel algorithm capable of finding parallel clusters in different subspaces and demonstrate our results for temporal and spatial applications. Our analysis of structural quality parameters in rivers is successfully used by hydrologists to developmeasures for river quality improvements.
Clustering Multidimensional Sequences in Spatial and Temporal Databases
2008
International Journal on Knowledge and Information Systems (KAIS) Vol. 16, Issue 1
data mining
29-51
dbtg