《计算机应用研究》|Application Research of Computers

基于分割图集的频繁闭图挖掘算法

Mining closed frequent graph based on partitioning graph database

免费全文下载 (已被下载 次)  
获取PDF全文
作者 薛冰,张俊峰,郑超
机构 1.河南城建学院 计算机科学与工程系,河南 平顶山 467036;2.燕山大学 信息科学与工程学院,河北 秦皇岛 066004
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2011)01-0061-04
DOI 10.3969/j.issn.1001-3695.2011.01.015
摘要 为了解决大规模图集挖掘算法PartGraphMining必须重复扫描图集才能得到全部频繁子图的缺点,提出了一种改进的IPMC算法,通过hash表保存同构图的hash地址和支持度,不必重复扫描图集就可快速得到全部频繁子图,再经过少量的子图同构判断得到全部频繁闭图。在实际数据集上运行的实验结果表明它比原算法的挖掘效率有所提高。
关键词 大规模图集;频繁子图;子图同构;频繁闭图
基金项目 国家自然科学基金资助项目(60673136);河南省重点科技攻关资助项目(092102210251)
本文URL http://www.arocmag.com/article/1001-3695(2011)01-0061-04.html
英文标题 Mining closed frequent graph based on partitioning graph database
作者英文名 XUE Bing, ZHANG Jun-feng, ZHENG Chao
机构英文名 1. Dept. of Computer Science & Engineering, Henan University of Urban Construction, Pingdingshan Henan 467036, China; 2. College of Information Science & Engineering, Yanshan University, Qinhuangdao Hebei 066004, China
英文摘要 In order to solve the shortage of the PartGraphMining algorithm for mining large-scale graph databases must repeatedly scan the database that could get all frequent subgraph patterns, this paper proposed a new algorithm IPMC.It could get all frequent subgraph patterns quickly without scanning the database repeatedly through storing graph’s hash address and supportting in the hash table.Furthermore, obtained all closed frequent graph patterns by the judgement of few subgraph isomorphism.The experimental result on real datasets shows that new algorithm improves the efficiency of mining.
英文关键词 large-scale graph databases; frequent subgraph patterns; subgraph isomorphism; closed frequent graph patterns
参考文献 查看稿件参考文献
 
收稿日期
修回日期
页码 61-64
中图分类号
文献标志码 A