摘要
运用关联规则对分布式数据库进行数据挖掘是一个常见的模式,为进一步提高在分布式挖掘多层关联规则算法的效率,改善内存的使用率,再次引入模糊理论和有效支持度的概念,并充分考虑有效支持度的阈值和有效支持度的支持频度,提出了一种新的产生频繁项集算法的修改方案,在理论上对此进行了分析和论证,实验证明这种算法的优化效果是明显的、是有用的。
In order to achieve better performance of the algorithm used to discovery multi-level association rule in distributed database, and also to get a desirable using rate of RAM, the theory of fuzzy and the conception of effective support are introduced in the paper, in addition, the researchers take into consideration both the frequency and the threshold of the effective support. The researchers propose an improved algorithm and give the result of experimentation and the theoretical analysis in detail to show the desirable performance enhancement being achieved.
出处
《浙江理工大学学报(自然科学版)》
2006年第1期50-55,共6页
Journal of Zhejiang Sci-Tech University(Natural Sciences)
基金
浙江理工大学科学基金项目(111251A4Y04002)
关键词
数据挖掘
频繁模糊项集
关联规则
有效支持度
Data mining
Frequent fuzzy itemsets
Association rules
Effective support