CAJ | 학술논문

关联规则挖掘是数据挖掘的重要领域之一,目前多数监督学习算法对满足最小支持度和最小置信度的关联规则进行深入分析的较少。剖析了分类关联规则挖掘算法CAR-Apriori算法,并提出了一种基于多最小支持度和支持度差别限制的分类关联规则挖掘算法MSCAR-Apriori算法。实验结果表明,改进算法不仅可以挖掘出满足给定条件的分类关联规则,同时还可以保留稀有但用户感兴趣且可能蕴涵巨大利润的规则项。
관련규칙알굴시수거알굴적중요영역지일,목전다수감독학습산법대만족최소지지도화최소치신도적관련규칙진행심입분석적교소。부석료분류관련규칙알굴산법CAR-Apriori산법,병제출료일충기우다최소지지도화지지도차별한제적분류관련규칙알굴산법MSCAR-Apriori산법。실험결과표명,개진산법불부가이알굴출만족급정조건적분류관련규칙,동시환가이보류희유단용호감흥취차가능온함거대리윤적규칙항。
Association rule mining is one of the important fields in Data mining,the most current association rules mining algorithms went less into deep analysis of association rules which meet minimum support and minimum confidence.This classification association rule mining algorithm i.e CAR-Apriori algorithm was analyzed,and based on the multiple minimum support and support difference constraint,an enhanced classification association rule mining algorithm MSCAR-Apriori algorithm was proposed.Experimental results expatiates that the improved algorithm can not only mine out the association rules to meet the given requirement,but also can keep the rule items rare,however,interested by users and also possibly implicates the rule item of huge profit.