DocumentCode :
1945553
Title :
The Traffic Accident Prediction Based on Neural Network
Author :
Fu Huilin ; Zhou Yucai
Author_Institution :
Sch. of Energy & Power Eng., Changsha Univ. of Sci. & Technol., Changsha, China
fYear :
2011
fDate :
5-7 Aug. 2011
Firstpage :
1349
Lastpage :
1350
Abstract :
The traffic accident prediction play an important role in the integrated planning and management of traffic, the reason which with much randomness about the traffic accident include some nonLinear elements, such as people, car, road, cLimate and so on. The traditional way of Linear analyses can not reveal the really situation since the noise pollution and amount of data are too Little, cause the result of prediction can not satisfactory. Because of the traditional BP network have some defects, such as local minimum, too many iterations, training too slow and so on, so choose improved LM-BP network to predict. Proposed a new way about the traffic accident prediction -- improved BP neural network, the traffic accident prediction have been analysis by the network.
Keywords :
accidents; backpropagation; neural nets; road safety; road traffic; BP neural network; integrated planning; neural network; nonlinear elements; traffic accident; traffic accident prediction; traffic management; Accidents; Economics; Educational institutions; Neurons; Pollution; Presses; Training; BP Neural Network; time sequence; traffic accident;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Digital Manufacturing and Automation (ICDMA), 2011 Second International Conference on
Conference_Location :
Zhangjiajie, Hunan
Print_ISBN :
978-1-4577-0755-1
Electronic_ISBN :
978-0-7695-4455-7
Type :
conf
DOI :
10.1109/ICDMA.2011.331
Filename :
6052126
Link To Document :
بازگشت