Railway Passenger Volume Forecasting Based on Support Vector Machine and Genetic Algorithm

Author

Chen, Xiaogang

Author_Institution

Digital Manuf. Technol. Lab., Huaiyin Inst. of Technol., Huaian, China

fYear

2009

fDate

6-7 June 2009

Firstpage

282

Lastpage

284

Abstract

A new prediction approach for the railway passenger volume is put forward by means of support vector machine optimized by genetic algorithm (GA-SVM). In GA-SVM model, GA is used to determine training parameters of support vector machine. GA has strong global search capability, which can get optimal solution in short time. Railway passenger volume of China from 1985-2002 is used to illustrate the performance of the proposed GA-SVM model. The experimental results indicate that the GA-SVM method can achieve greater forecasting accuracy than artificial neural network in railway passenger volume forecasting.

Keywords

forecasting theory; genetic algorithms; rail traffic; support vector machines; traffic engineering computing; genetic algorithm; railway passenger volume forecasting; support vector machine; Artificial neural networks; Computer aided manufacturing; Fault tolerance; Genetic algorithms; Learning systems; Parallel processing; Rail transportation; Statistical learning; Support vector machines; Technology forecasting; forecasting method; genetic algorithm; railway passenger volume; support vector machine;

fLanguage

English

Publisher

ieee

Conference_Titel

Future Computer and Communication, 2009. FCC '09. International Conference on

Conference_Location

Wuhan

Print_ISBN

978-0-7695-3676-7

Type

conf

DOI

10.1109/FCC.2009.81

Filename

5235649