基于概率衰减窗口模型的不确定数据流频繁模式挖掘.pdfVIP

ISSN 1 计算机研究与发展 1000—1239／CN1—1777／TP of Researchand JournalCompuler Development 基于概率衰减窗口模型的不确定数据流频繁模式挖掘 廖国琼“2 吴凌琴3 万常选1’2 1(江西财经大学信息管理学院南昌 330032) 2(江西省高校数据与知识工程重点实验室南昌330032) 3(江西省赣抚平原水利工程管理局南昌 330201) (1iaoguoqiong@163．corn) Patterns overUncertainDataStreamsBasedon Frequent Mining WindowModel Decay Liao Wan Guoqion91”，WuLingqin3，andChangxuanl’2 FinanceandEconomicsj 330032) 1(School。，InformationTechnology，JiangxiUniversityof Nanchang Dataand 330032) 2(]iangxiKeyLaboratoryof KnowledgeEngineering，Nanchang Water AdministrationBureau 330201) Plain，Nanchang 3(JiangxiConservancyProject ofCanfu to AbstractInrecent amountsofuncertaindataare due thewide of years，alarge emerging usage new aswirelesssensornetworksandradio the such technologies frequency ofuncertaindatastreams，anewkindof tree—PFP。treeanda uncertainty probabilityfrequentpattern are inthis probabilityfrequentpatternminingmethod--PFP—growthproposed usestransactionaluncertaindatastreammodelanda time—based windowmodelto probabilitydecay main find characteristicsof probabilityfrequentpatternsthroughcalculatingexpectedsupports．The include：1)Becausethecontributionsonthe ofitems at PFP—g