基于相似關(guān)系的粗集理論及其在KDD中的應(yīng)用.pdf

1、西南交通大學(xué)碩士學(xué)位論文基于相似關(guān)系的粗集理論及其在KDD中的應(yīng)用姓名：劉仰光申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：應(yīng)用數(shù)學(xué)指導(dǎo)教師：吳建樂2001.4.1西南交通大學(xué)研究生學(xué)位論文第ii頁(yè)AbstractInthispap％theclassicalrou【ghsetsareextendedTheclassicalroughsetsarebasedOntheindiscerniblerelationieequivalencerelationButin

2、fact，thisconditionisn’teasytobesatisfiedintherealappHcafions，fortherealdataisincompleteSimilarityrelationgenerallyexistsinthereallife，SOitisnatul鋤tousethesimilarityrelationtotaketheplaceoftheindiscerniblerelationbeinguse

3、dintheclassicalroughsetsInthispaperwefocusonthedefinitionsandpropertiesoftheextendedroughsetsbasedorisimilarityThedifferencesbetweenclassicalandextendedroulghsetsarediscussedBesides，discussingthenumbercharacterishelpfult

4、othepracticalapplicationsSOtheroughmembershipfunctionbasedonsimilarityisdefmedandsomepropertiesareprovedDuringwhich，theupperapproximationandlowapproximationbasedOilsimilarityareredefined，butbyprovingthepropertiesoftherou

5、ghmembershipfunctionbasedonsimilaritythetwodefinitionsareequalOntheotherhanditisachallengingworktoapplytheroughsetsInthispaperwediscussthedataminingbasedonroughsetsFirstthemethodofdataminingbasedonroughsetsisoneofthevari

6、ousdataminingm甜tlodsIthasthegeneralcharactersbeingthedataminingmethodsThenitisar刪ldndofdataminingmethodwhichhasmoreadvantagescomparedtotheothea鴨egdatareductionThismethodusuallyinehdesthefollowingmainstq3s：dataproproceedi