J4 ›› 2010, Vol. 28 ›› Issue (03): 264-.

• 论文 • 上一篇    下一篇

关联规则算法的电信网络告警应用

于 漫1,胡 明1,金 刚2,胡 亮2,赵 阔2
  

  1. 1.长春工业大学 计算机科学与工程学院|长春130012;2.吉林大学 计算机科学与技术学院|长春 130012
  • 出版日期:2010-05-30 发布日期:2010-06-12
  • 通讯作者: 于漫(1984— ),女,长春人,长春工业大学硕士研究生,主要从事数据挖掘研究,(Tel)86-13514470553 E-mail:237055165@qq.com
  • 作者简介:于漫(1984— )|女|长春人|长春工业大学硕士研究生|主要从事数据挖掘研究|(Tel)86-13514470553(E-mail)237055165@qq.com;胡明(1963— )|男|长春人|长春工业大学教授|硕士生导师|主要从事人工智能和数据挖掘研究|(Tel)86-13331757733(E-mail)huming@mail.ccut.edu.cn
  • 基金资助:

    吉林省科技厅基金资助项目(20060305)

Association Rules Algorithm Applied to Telecommunications
Network Alarms

YU Man1, HU Ming1, JIN Gang2, HU Liang2, ZHAO Kuo2
  

  1. 1.College of Computer Science and Engineering, Changchun University of Technology, Changchun 130012, China;2. College of Computer Science and Technology, Changchun Jilin University, Changchun 130012,China
  • Online:2010-05-30 Published:2010-06-12

摘要:

为提高网络告警数据的处理效率,迅速找出故障根源,以保证整个电信网络的正常运行。针对告警数据的特点对传统的关联规则算法进行了相应的改进,提供了优先剪枝策略,避免了生成没有意义的频繁项集;k-1-项集通过连接操作生成k项集,避免了反复扫描数据库所带来的压力。结果表明,改进算法提高了挖掘效率,达到快速诊断网络故障的目的。

关键词: 告警诊断, 数据挖掘, 关联规则, Apriori算法

Abstract:

Modern telecommunications networks are very powerful, and there is a sharp increase in business volume, and network is becoming large, complex, heterogeneous. In order to ensure the normal operation of the entire telecommunications network, we must improve the efficiency of processing network alarm, and identify the root causes of failure rapidly. Association rules algorithm of data mining can effectively analyze the relevance of things happened in the network warning system. It can also accurately locate the failure of a network application.We improved the traditional association rules algorithm corresponding the characteristics of alarm data. We improved mining efficiency and achieved a rapid diagnosis of network failures.

Key words: alarm diagnostic, data mining, association rules, Apriori algorithm

中图分类号: 

  • TP311