摘要
为了提高关联规则挖掘算法处理大数据集的能力,在基因表达式编程进化算法(Gene Expression Program-ming)的基础上,提出了一个新的挖掘强关联规则的算法框架。主要贡献在于提出并实现了基于小生境技术的基因表达式编程进化算法NGEP,以用于挖掘关联规则。NEGP算法首先进行小生境演化,融合小生境并剔除同构的优秀个体,然后对小生境解进行笛卡儿交叉,以产生更好的结果。实验结果表明,与同类优秀的算法对比,NGEP算法的种群多样性与精确度都有很好的结果,并且在提取有效规则的效率上也有较大的提高。
In order to advance the performance of association rules mining algorithm when disposing big dataset,a novel GEP based on Niche （NGEP） was presented to solve the probleme. The procedure of NGEP began with the niche evolution and then fused some of the sub-niches according to the similarity of best individuals. Then Cartesian product was nested in the kernel set of niches to generate better outcomes. The experimental results show that our algorithm performs better than the other similar evolutionary algorithm in terms of diversity of population and precision; besides, it can discover more association rules.
出处
《计算机科学》
CSCD
北大核心
2009年第11期224-227,共4页
Computer Science
基金
国家自然科学基金资助项目(No.60603074和No.60603075)
国家973重大研究项目(2004CB318203)
中国地质大学(武汉)优秀青年教师资助计划项目(CUGQNL44)资助
关键词
数据挖掘
关联规则
基因表达式编程
小生境
Data mining, Association rule,Gene expression programming, Niche
