摘要
基于模糊集理论,挖掘带有数量属性的序列模式称为模糊序列模式挖掘。源于AprioriAll算法的模糊序列模式挖掘算法需多次扫描数据库。针对该缺点,提出一种基于序列矩阵表示且只需扫描一次数据库的算法MFSPM。实验表明,算法效率有明显提高。
It is called fuzzy sequential patterns mining that we mine sequential patterns with quantities based on fuzzy sets theory.The algorithm originated from AprioriAll needs to scan the database for several times.Against the disadvantage,proposes a new algorithm MFSPM that it adopts a matrix to express a sequence and only needs to scan database once.Experiment shows that it is much more efficient.
出处
《现代计算机》
2010年第11期53-57,共5页
Modern Computer
关键词
数据挖掘
模糊集
模糊序列模式
矩阵
Data Mining
Fuzzy Sets
Fuzzy Sequential Patterns
Matrix