数据挖掘取样方法与研究_胡文瑜.pdfVIP

ISSN 1000-1239 CN 11-1777 TP Journal of Computer Research and Development 8(1): 5-5 , 2011 1, 2 1 1, 3 胡文瑜 孙志挥 吴英杰 　 　 1 (　　210096) 2 (　　350108) 3 (　　350108) (huw enyu @) Study of Sampling Methods on Data Mining and Stream Mining 1, 2 1 1, 3 Hu Wenyu , Sun Zhihui , Wu Yingjie 1 (School of Comp uter S cience and Engineering, Southeast University, Nanj ing 210096) 2 (Dep artment of Comp uter and I nf ormation Science, F uj ian University of Technology, Fuzhou 350108) 3 (College of Mathematics and Comp uter Science, F uz hou University, F uz hou 350108) Abstract　Sampling is an efficient and most w idely-used approximation technique.It enables lots of algorithms to be applied to huge dataset by use of scaling dow n dramatically dataset for data mining and streaming mining .T hroughout the detailed review, a kind of taxonomic frame of sampling algorithms based on uniform sampling and biased sampling is presented;meanw hile, analysis, comparisons and evaluations of representative sampling algorithms such as reservoir sampling, concise sampling, count sampling, chain-sampling, DV sampling and so on are performed.Due to the limitations of uniform sampling in some applications—queries w ith relatively low selectivity, outlier detection in large multidimensional data sets, and clustering over data streams w ith skew ed Zipf distribution, the importance of need for using biased sampling methods in these scenarios is fully dissertated.In addition to listing successful applications of sampling techniques in data mining, statistics estimating and stream mining up to now, w e survey the application and development of sampling techniques, especially those traditional classic sampling techniques such as progressive sampling, adaptive sam