An Improved Support Vector Machine Algorithm and its Application in Intelligent Transportation System

Intelligent transportation system as the combination of computer technology, information technology, communication technology, electronic control technology and transportation has been a hot point has been widely used to save many existing problems in transportation. It is widely accepted that traffic incident has strong randomness and unpredictable destructiveness Support vector machine proposed by Vapnik et al. is introduced in this paper to solve the existing problems in traffic incidents in order to help improve the efficiency and effect in intelligent transportation system. Here, support vector machine is improved by introducing particle swarm optimization (PSO) which is a powerful and easy way to implement. This improved method can optimize both the optimal feature subset and parameters in SVM, which can further to solve time in computation. Finally, an experiment is demonstrated to show the application of the proposed method in intelligent transportation system.