基于多支持度的正負關(guān)聯(lián)規(guī)則挖掘技術(shù)的研究.pdf

1、山東輕工業(yè)學(xué)院碩士學(xué)位論文基于多支持度的正負關(guān)聯(lián)規(guī)則挖掘技術(shù)的研究姓名：李剛申請學(xué)位級別：碩士專業(yè)：計算機應(yīng)用技術(shù)指導(dǎo)教師：王洪國20080503ABSTRACT——_一一——————————————————————————————————_—————_——————————_————————一ABSTRACTAssociationrule(AR)isoneofthemostimportanttechniquesofdatamining，

2、theexistingassociationrulesminingalgorithmsaremostbasedonfrequentitemsets，theresearchsaboutinfrequentitemsetsareverylittleHoweverwhenwestudythenegativeassociationrules，infrequentitemsetsbecomemoreandmoreimportantbecauset

3、heycontainalotimportantnegativeassociationrulesAtthesametime，theexistingalgorithmsofminingassociationrulesarealmostachievedbythesingleminimumsupportbutinfacttheoccurrencefrequenciesoftheeventscorrespondingtoeachattribute

4、areverydifferent，SOtherestrictionofsingleminimumsupportisunableteflecttheobjectiveForthesakeofoffsetingtheshortageofsingleminimumsupport，multipleminimumsupportalgorithmsappearThisarticlebringsforwardthreenewalgorithmsbas

5、edonmultipleminimumsupportalgorithmswhichameliorateandperfecttheoldmultipleminimumsupportalgorithmsThethreenewalgorithmsareMMS_inFSalgorithm，2LMSinFSFSalgorithmandMLMSalgorithmTheMMSinFSalgorithmaddsarestrictionofinfrequ

6、entitemsetsbasedonmultiplesupportsalgorithmwhichmakesitcanminesbothinfrequentitemsetsandfrequentitemsetssimultaneousThe2LMSinFS—FSalgorithmisbasedontwolevelmultiplesupports，itamelioratestheMMS—inFSalgorithmwhichmakesares

7、trictionofinfrequentitemsetsfromsinglesupporttomultiplesupportsItsetstwolevelsupportsforeachitem，oneisusedtestrictinfrequentitemsetsandtheotherisusedtestrictfrequentitemsetsInthiswayitCanrestrictfrequentitemsetsandinfreq

8、uentitemsetsbetterTheMLMSalgorithmisbasedonmultiplelayerminimumsupportswhichsetsdifferentminimumsupportsfordifferentitemsForthenewalgorithms，wecomparedthemwithseveralotheralgorithmmodelsandwedosomeexperimentionstoproveth