CAJ | 학술논문

关联规则挖掘是数据挖掘中的一个重要研究内容.为了高效、快速地从事务数据库中挖掘出频繁项集,针对数据挖掘的经典关联规则Apriori算法的瓶颈问题提出了改进的方法.算法将事物数据库映射到布尔型数组中,然后所有的操作都针对数组元素值展开.这样大大减少了数据库的扫描次数.算法利用数组的随机访问特性及布尔型数据的简单"与"操作,直接产生频繁项集,而不产生大量的候选项集.经理论分析和实验结果显示该算法在效率上明显优于Apriori 算法.
관련규칙알굴시수거알굴중적일개중요연구내용.위료고효、쾌속지종사무수거고중알굴출빈번항집,침대수거알굴적경전관련규칙Apriori산법적병경문제제출료개진적방법.산법장사물수거고영사도포이형수조중,연후소유적조작도침대수조원소치전개.저양대대감소료수거고적소묘차수.산법이용수조적수궤방문특성급포이형수거적간단"여"조작,직접산생빈번항집,이불산생대량적후선항집.경이론분석화실험결과현시해산법재효솔상명현우우Apriori 산법.
Association rule mining is an important part of research content in data mining.In order to efficiently and quickly mine all frequent iternset from the transaction database,an improed algorithm of mining association rules is presented for the bottleneck problem of the classic Apriori algorithm.The transaction database is mapped to Bool array,then all the operations are carried out based on array elements value,thereby reducing the database scanning frequency.Then use bitwlse "AND" operation and random access characteristics of array,a direct consequence of frequent itemsets,rather than have a large nurnber of candidate sets.thereby improving the efficiency of the algorithm.