现代电子技术Issue(20)：134-136,140,4.
基于决策树算法的工业光纤网络通信故障检测系统
Industrial fiber network communication fault detection system based on decision tree algorithm
摘要
Abstract
For tedious fault detection process and low detection efficiency of industrial optical network communication,an optical fiber network communication fault detection system based on optimized decision tree data mining algorithm was designed. In order to reduce the repeated rework of detector and realize accurate positioning,the fault data mining module taking the deci-sion tree as the core was introduced into the system to reduce the time of fault mining in not its own space. Only taking the IP address as a parameter in the first-order fiber-optic network fault classification is changed into the construction of decision tree based on the specified network index in the second-order classification,which further improves the accuracy of fault detection. The application results of optical network existing in a vehicle manufacturing enterprise show that,compared with the standard method of decision tree,this algorithm has increased accuracy from 69.0% to 99.9%,and reduced the false report rate from 3.14% to 0.48%. The optimization effect is obvious.关键词
光纤网络/通信故障检测/故障诊断/数据挖掘/决策树Key words
fiber network/communication fault detection/fault diagnosis/data mining/decision tree分类
信息技术与安全科学引用本文复制引用
黄勇,胡连城,刘增良..基于决策树算法的工业光纤网络通信故障检测系统[J].现代电子技术,2014,(20):134-136,140,4.基金项目
国家自然科学基金（90818025）；国家自然科学基金 （）
