The constructions of Haar wavelet synopses for large datasets have proven to be useful tools for data approximation. Recently, research on constructing wavelet synopses with a guaranteed maximum error has gained attention. The goal is to find optimal synopses that minimize the approximation error under certain metrics. There are two approaches to realize this goal: size bounded and error bounded. In this paper, we provide a new algorithm for building error-bounded synopses. Our approach is based on the heuristic approach, achieving near-optimal accuracy and superior runtime performance. In addition, we provide the pruning techniques, which can greatly improve the performance of error bounded synopses construction. We then demonstrate the effectiveness of our algorithm through extensive experiments.

No abstract