This paper presents a compression method, PatZip, to improve the efficiency of spatial pattern mining methods. PatZip can avoid overcompression and stop automatically before pattern is destroyed. Compared with existing compression methods, PatZip is deterministic and its result is reproducible, and original data can be easily recovered. The compression process is data-driven and parameter-free, and requires only O(nlogn) time for n data points. © Springer-Verlag Berlin Heidelberg 2005.
CITATION STYLE
Qian, Y., Zhang, K., & Huynh, D. T. (2005). PatZip: Pattern-preserved spatial data compression. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 3518 LNAI, pp. 726–736). Springer Verlag. https://doi.org/10.1007/11430919_84
Mendeley helps you to discover research relevant for your work.