摘要
提出一种基于粗糙集理论和布尔矩阵的关联规则挖掘算法,作为对Apriori算法的改进,通过构造布尔矩阵,利用粗糙集划分等价类的方法对事务数据库的记录进行分类,然后通过等价类的取交或取并运算产生更高阶的频繁项目集,算法能有效减少数据库的扫描次数,实验表明算法在对事务数据库进行挖掘时显示出良好的性能.
This paper proposes an association rule mining algorithm based on a rough set theory and Boolean matrix as the improvement of the Apriori algorithm, which classifies the records of transaction database by constructing a Boolean matrix. The rough set division method is used to obtain equivalence classes, and then the equivalence class helps generate the more advanced itemsets. The algorithm can effectively reduce the number of database scanning. Experiment results demonstrate that when the transaction database meets certain conditions, the algorithm shows a good performance.
出处
《西南民族大学学报（自然科学版）》
CAS
2013年第3期366-372,共7页
Journal of Southwest Minzu University(Natural Science Edition)
