交通运输系统工程与信息

交通运输系统工程与信息 ›› 2004, Vol. 4 ›› Issue (4): 54-59 .

• 系统工程理论与方法 • 上一篇    下一篇

基于粗集理论预处理数据的神经
网络交通事件自动检测算法

温娟，贺国光   

  1. 天津大学系统工程研究所，天津300072
  • 收稿日期:2004-04-30 修回日期:1900-01-01 出版日期:2004-11-01 发布日期:2004-11-01

Automatic Detection Algorithm of Traffic Incident
Using Neural Network Based on Rough Sets Filtration

WEN Juan，HE Guo-guang   

  1. Institute of Systems Engineering, Tianjing University, Tianjin 300072, China
  • Received:2004-04-30 Revised:1900-01-01 Online:2004-11-01 Published:2004-11-01

PDF

摘要： 提出了一种基于粗集理论预处理数据的神经网络交通事件自动检测算法.首先简单介绍T数据挖掘技术中的粗集理论，然后说明了算法的原理、计算步骤以及所用的神经网络模型，最后为检验算法的有效性，时MATLAB编制的仿真程序的生成试验数据做了较充分的仿真试验.试验测试结果表明了此方法的有效性.

关键词: 交通事件, 自动检测算法, 粗集理论, 神经网络

Abstract: Based on rough sets filtration, an automatic detection algorithm of traffic incident using neural network is presented. After briefly introducing of Rough Sets in data mining technology, the principle and computing process of the algorithm and the neural network model which was used in this paper are explained. In order to test the validity of the algorithm, enough tests have been done using the testing data obtained from the simulation programmed by MATLAB. Both the testing process and the results of the simulation are introduced clearly. The results show that the algorithm is effective.

Key words: traffic incident, automaticd etectiona lgorithm, rough sets, neural network