University of North Alabama
Data streams are often locally correlated, with a subset of streams exhibiting coherent patterns over a subset of time points. Subspace clustering can discover clusters of objects in different subspaces. However, traditional subspace clustering algorithms for static data sets are not readily used for incremental clustering, and are very expensive for frequent re-clustering over dynamically changing stream data. In this paper, the authors present an efficient incremental subspace clustering algorithm for multiple streams over sliding windows.