基于改进型BP神经网络的随机车祸持续时间预测

由于道路交通事故的复杂性使之对事故持续时间的预测困难。采用因子分析和BP神经网络相结合的方法，以福银高速福州段近两年交通数据为依据，采用因子分析获取造成车祸事故的少量公共因子，将公共因子做降维处理后作为BP神经网络的输入参数，利用三层BP神经网络实现对随机车祸持续时间的预测。其结果与回归算法、支持向量机算法以及传统BP神经网络算法相比，精准度高、收敛速度快。

Prediction of Random Traffic Accident Duration Based on Improved BP Neural Network

Due to the complexity of road traffic accidents, it is difficult to predict the accident duration. Based on the traffic data of Fuzhou section of Fuzhou Yinchuan Expressway in recent two years, factor analysis is used to obtain a small number of public factors causing traffic accidents, and the public factors are used as the input parameters of BP neural network after dimensionality reduction. The three-layer BP neural network is used to predict the duration of random traffic accident. Compared with regression algorithm, support vector machine algorithm and traditional BP neural network algorithm, the result has high accuracy and fast convergence.