关联规则挖掘Apriori算法的研究与改进  

Research and An improvement on Apriori Algorithm based on mining association rules

作　　者：王铮[1] 周国光[1] 

机构地区：[1]重庆大学计算机学院,重庆400044

出　　处：《网络安全技术与应用》2011年第4期61-62,共2页Network Security Technology & Application

摘　　要：本文采用一种基于布尔矩阵的频繁集挖掘算法。该算法直接通过支持矩阵行向量的按位与运算来找出频繁集,而不需要Apriori算法的连接和剪枝,通过不断压缩支持矩阵,不仅节约了存储空间,还提高了算法的效率。The discovery of association rules in datamining is an important issue,the core of which is the frequent pattern mining,Apriori algorithm is classical for the association rule mining,but it should repeatedly scan thed database and can produce plenty of candidates. The paper proposed a new algorithm based on Boolean matrix ,Through direct support of the matrix by the vector-line operations and to identify frequent sets ,the new algorithm simplifies the join step and the prune step in Apriori algorithm.Also it reduces the support matrix unceasingly to reduce the higher mode frequent collection excavation time and save the storage space.

关 键 词：数据挖掘 关联规则 APRIORI算法 频繁项集 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]