Building Data Synopses Within a Known Maximum Error Bound

The constructions of Haar wavelet synopses for large data sets have proven to be useful tools for data approximation. Recently, research on constructing wavelet synopses with a guaranteed maximum error has gained attention. Two relevant problems have been proposed: One is the size bounded problem that requires the construction of a synopsis of a given size to minimize the maximum error. Another is the error bounded problem that requires a minimum sized synopsis be built to satisfy a given error bound. The optimum algorithms for these two problems take O(N2 ) time complexity. In this paper, we provide new algorithms for building error-bounded synopses. We first provide several property-based pruning techniques, which can greatly improve the performance of optimal error bounded synopses construction. We then demonstrate the efficiencies and effectiveness of our techniques through extensive experiments.

• Type: Conference
• Year: 2007
• Where: WAIM
• Authors: Chaoyi Pang, Qing Zhang, David P. Hansen, Anthony J. Maeder
• Keywords: Haar Wavelet Synopses, Information Management, Maximum Error, WAIM 2007, Wavelet Synopses