作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2012, Vol. 38 ›› Issue (7): 31-33. doi: 10.3969/j.issn.1000-3428.2012.07.011

• 软件技术与数据库 • 上一篇    下一篇

一种频繁子图挖掘算法

唐德权1,谭 阳2   

  1. (1. 湖南警察学院信息技术系,长沙 410138;2. 湖南广播电视大学信息技术系,长沙 410004)
  • 收稿日期:2011-08-01 出版日期:2012-04-05 发布日期:2012-04-05
  • 作者简介:唐德权(1979-),男,讲师、硕士,主研方向:信息安 全,数据挖掘;谭 阳,讲师、硕士
  • 基金资助:
    湖南省教育厅科研基金资助项目(10C0134);湖南省自然科学基金资助项目(06JJ50107);湖南省教育厅基金资助重点项目 (10A074)

Algorithm for Frequent Subgraph Mining

TANG De-quan 1, TAN Yang 2   

  1. (1. Department of Information Technology, Hunan Police Academy, Changsha 410138, China; 2. Department of Information Technology, Hunan Radio & TV University, Changsha 410004, China)
  • Received:2011-08-01 Online:2012-04-05 Published:2012-04-05

摘要: 为减少频繁子图规范化检测的时间复杂度,对规范化邻接矩阵的相关性质进行分析。给出相关定理并证明其正确性,从而减少冗余候选子图的产生。在此基础上,提出一种频繁子图挖掘算法——FSM_CAM。实验结果证明,与现有频繁子图挖掘算法FSubGraphM相比,FSM_CAM算法的效率较高。

关键词: 频繁子图, 规范邻接矩阵, 候选子图, 数据挖掘

Abstract: To reduce the frequent subgraph canonical detection time complexity, this paper discusses the relevant properties of the canonical adjacency matrix. In order to reduce the generation of redundant candidate subgraphs, it proposes the theorem to reduce redundancy candidate subgraph generation, and proves its correctness. Based on this, it proposes a frequent subgraph mining algorithms called FSM_CAM. Experimental results show that FSM_CAM algorithm’s efficiency is greatly improved compared with existing frequent subgraph mining algorithms FsubGraphM.

Key words: frequent subgraph, Canonical Adjacency Matrix(CAM), candidate subgraph, data mining

中图分类号: