多维频繁情节挖掘在电信告警信息分析中的应用

电信网络每天都要产生大量的告警信息,这些信息中隐藏着网络结构相关的有用知识。基于对电信网络告警信息的特点的分析和针对现有挖掘方式的不足,论文提出一种从电信网络告警信息中挖掘频发的模式知识的思想方法———多维频繁情节挖掘。挖掘的多维频繁情节可以帮助网络管理人员分析告警信息和诊断故障。

MDEM for Analysis of Telecommunication Network Alarms

Telecommunication networks produce large amounts of alarm information daily.This data contains potentially valuable knowledge about the network.Based on characteristics of alarm information of telecommunication network and to the deficient of nowadays algorithms ,the authors present a knowledge discovery methodology,Multi-Dimension Frequent Episodes Mining(MDEM),to mine recurrent alarm patterns from telecommunication network alarm information databases.The discovered patterns can be employed to help network managers analyze alarms and locate the faults.