AprioriAll和GSP算法的研究及实现

陕西理工学院毕业论文AprioriAll和GSP算法的研究及实现作者：刘剑青（陕西理工学院数学系信息与计算科学专业044班陕西723000）指导教师：周涛[摘要]：序列模式挖掘即从序列数据库中发现频繁子序列以作为模式,它是一类重要的数据挖掘问题,有着非常广泛的应用前景,被应用在包括顾客购买行为的分析、网络访问模式分析、科学实验的分析、疾病治疗的早期诊断、自然灾害的预测、DNA序列的破译等方面。的效率上。本文我就是在对序列模式挖掘的其中两种算法进行研究，即：Armorial和GSP算法。首先讲述了序列模式的一些概念及基本原理。并通过具体的实例演示算法的执行过程，进而达到掌握理解的成度。再次基础上采用vc编程语言和Access数据库进行实现，最后对程序运行结果进行分析和总结。关键字：数据挖掘Armorial算法GSP算法陕西理工学院毕业论文TwoDataMiningAlgorithmsAuthor：LiuJianqing（Grade04,Class04,Informationandcalculationscience，DepartmentofMathematics，ShaanxiUniversityofTechnology，Hanzhong723000，Shaanxi）Tutor:ZhouTaoAbstract:Sequentialpatternminingfromthesequencefoundinthedatabaseasasequenceoffrequentpattern,itisakindofimportantdataminingissues,hasaverywideapplication,beusedincustomerbuyingbehavior,includingtheanalysisofnetworkaccessmodeofanalysis,thescientificexperimentsAnalysis,theearlydiagnosisofdisease,naturaldisastersforecast,DNAsequencesdeciphered,andsoon.Theefficiency.Inthispaper,Iwasinthesequencepatternminingoneoftwoalgorithmstostudy,namely:ArmorialandGSPalgorithm.Firstonthesequencepatternsofsomebasicconceptsandprinciples.Anddemonstratethroughconcreteexamplesoftheimplementationofthealgorithm,thenreachedintothegraspofunderstanding.UsedvcagainbasedontheprogramminglanguageandAccessdatabasetoachievetheendresultofrunningtheanalysisandsynthesis.Keyword:DataminingalgorithmAprioriallalgorithmGSPAlgorithm陕西理工学院毕业论文目录１．序列模式挖掘的基本概念..................................................................................................-1-2．AprioriAll算法学习...........................................................................................................-2-2．1基本思想.....................................................................................................................-2-2．2AprioriAll算法的基本思路.......................................................................................-2-2．3应用举例....................................................................................................................-2-2．4AprioriAll算法程序实现.......................................................................................-5-2．4．1AprioriAll算法的描述.............................................................................-5-2．4．2AprioriAll算法的程序.............................................................................-6-2．4．3AprioriAll算法程序的结果.............................................................................-6-2．4．4AprioriAll算法程序运行结果分析.......................................................-7-2．5AprioriAll算法的本质...........................................................................................-7-2．6AprioriAll算法存在的问题..................................................................................-7-3．GSP算法的学习....................................................................................................................-8-3．1GSP算法的基本思想................................................................................................-8-3．2GSP算法的基本思路................................................................................................-8-3．3产生候选序列模式的步骤........................................................................................-8-3．4后选集计数................................................................................................................-9-3．5GSP算法程序实现..................................................................................................-10-3．5．1GSP算法的程序......................................................................................-10-3．5．2GSP算法程序结果..................................................................................-10-3．5．4GSP算法程序运行结果分析.....................................................................-11-3．6GSP算法分析..........................................................................................................-12-3．7GSP法存在的问题.................................................................................................-12-4．对于Apriori的算法小结.................................................................................................-12-5．研究发展展望......................................................................................................................-12-致谢...........................................................................................................................................-12-参考文献....................................................................................................................................-13-附件一：....................................................................................................................................-14-附件二.......................................................................................................................................-31-陕西理工学院毕业论文-1-１．序列模式挖掘的基本概念项目集或称项集(Itemset)是各种项目组成的集合。设I={i1,i2,…,im}是一个项目集合,事务数据库D={t1,t2,…tn}是由一系列具有惟一标识TID的事务组成,每个事务ti(i=1,2,…,n)都对应I上的一个子集。设I1⊆I,项目集I1在D上的支持度(support)是包含I1的事务在D中所占的百分比,即support(I1)=||{t∈D|I1⊆t}||/||D||对项目集I和事务数据库D,I中所有满足用户指定的最小支持(Minsupport)的项目集,即大于或等于Minsupport