摘要
在对关联规则中的Apriori算法进行了深入研究的基础上,提出了基于矩阵结构的关联规则挖掘算法。由于这个算法只需要对交易数据库进行一次搜索,给出了一种简单有效的逐步缩减交易数据库的方法,能大量减少所需的I/O次数,因此提高了Apriori算法的效率,并改进了数据挖掘算法的性能。
Based on fully analyzing the Apriori algorithm of an association rule mining algorithm, this paper presents a new matrix - based association rule mining algorithm. This improved algorithm scans the database only once and a highly efficient method of minimizing the trade database is given. So it reduces the times of input and output, thus the new algorithm promotes the algorithm efficiency and at the same time improves the performance of the data mining technique.
出处
《计算机与现代化》
2007年第7期3-5,共3页
Computer and Modernization
关键词
数据挖掘
关联规则
事务压缩
频繁项集
data mining
association rule
transaction reduction
frequent item-set