A Semi-Random Multiple Decision-Tree Algorithm for Mining Data Streams-一种用于挖掘数据流的半随机多决策树算法.pdfVIP

XDeta1．tAsemi-randomdecision-treefor data OF Hu mining XG，LiPP，Wu multiple algorithm COMPUTERSCIENCEANDTECHNOLOGY 22(5)：711—724Sept．2007 ASemi--RandomDecision—-Treefor Data Multiple AlgorithmMining Streams Gong—QingWul(吴共庆) Xue—GangHul(胡学钢)，Pei．PeiLil(李培培)，Xin-DongWul，2(吴信东)，and 1School Scienceand o|Computer InformationEngineering}HefetUniversityofTechnology，Hefei230009，China ofComputerScience，UniversityofVermont，Burlington，VT50405,U．S．A． 2Department Received January15，2007；revisedJuly16，2007 data with dataisahot indata classificationon streams， Abstract performing Miningstreaming topic mining．When classificationbasedondecision asID3and a inboth traditional algorithms trees，such C4．5，haverelativelypoore伍ciency and when duethe of data．Therearesome intime timeand to characteristics advantages space using space streaming for data decision randomdecisiontrees．Anincremental Multiple algorithmmining treesis inthis usesthe of TreesforData onrandomdecision inequality Streams)，based proposedpaper．SRMTDS heuristicme