基于磁盘表存储FP-TREE的关联规则挖掘算法.pdf

ISSN 计算机研究与发展 IOOO-1239／CN11—1777／TP of Researchand JournalComputer Development 基于磁盘表存储FP-TREE的关联规则挖掘算法 申 彦h2 宋顺林2 朱玉全2 1(江苏大学信息管理与信息系统系江苏镇江212013) 2(汀苏大学计算机科学与通信工程系江苏镇江212013) 104186179@qq．corn) ofAssociationRulesBasedonDiskTableResidentFP—TREE MiningAlgorithm Shen Shunlinz，andZhu Yanl一，Song Yuquan2 and 212013) 1(DepartmentofInformationManagementIncyormationSystem，Jiangsu ScienceandCommunication 212013) 2(DepartmentoyComputer Engineering·Jiangsu AbstractAsthesizeofthedatabasetobeminedis sizeof increasingconstantly，the memory physical availablehasbecomeabottleneckwhen FP-GROWTHforassociationrules using algorithm mining． So。itis totackle somenew inordertomineassociationrules necessary spacescalabilityby algorithms in hasbecomethemain database．Nowadays，disk—resident huge algorithm datastructureofFP—TREEis andanovel called original improved algorithm tableresidentFP-TREE usesdisktablefor FP-TREEto algorithm growth)ispresented．This storing failedFP—GROWTHtoo decrease the works for much memoryusage．Whenmining using memory， DTRFPGROWTHcancontinuetomineassociationrulesfrom databaseits skillcalled huge by special disktableresidentFP-TREE，whichis suitab