摘要
Approximatequeryprocessinghasemergedasanapproachtodealingwiththehugedatavolumeandcomplexqueriesintheenvironmentofdatawarehouse.Inthispaper,wepresentanovelmethodthatprovidesapproximateanswerstoOLAPqueries.Ourmethodisbasedonbuildingacompressed(approximate)datacubebyaclusteringtechniqueandusingthiscompresseddatacubetoprovideanswerstoqueriesdirectly,soitimprovestheperformanceofthequeries.WealsoprovidethealgorithmoftheOLAPqueriesandtheconfidenceintervalsofqueryresults.AnextensiveexperimentalstudywiththeOLAPcouncilbenchmarkshowstheeffectivenessandscalabilityofourcluster-basedapproachcomparedtosampling.
出版日期
2002年05月15日(中国期刊网平台首次上网日期,不代表论文的发表时间)