Abstract
To increase the efficiency and precision of large-scale road network traffic flow prediction, a genetic algorithm-support vector machine (GA-SVM) model based on cloud computing is proposed in this paper, which is based on the analysis of the characteristics and defects of genetic algorithm and support vector machine. In cloud computing environment, firstly, SVM parameters are optimized by the parallel genetic algorithm, and then this optimized parallel SVM model is used to predict traffic flow. On the basis of the traffic flow data of Haizhu District in Guangzhou City, the proposed model was verified and compared with the serial GA-SVM model and parallel GA-SVM model based on MPI (message passing interface). The results demonstrate that the parallel GA-SVM model based on cloud computing has higher prediction accuracy, shorter running time, and higher speedup.
Highlights
Large-scale road network has high complexity, strong nonlinear, and high dynamic
The results demonstrate that the parallel genetic algorithm-support vector machine (GA-SVM) model based on cloud computing has higher prediction accuracy, shorter running time, and higher speedup
Wang proposed a parallel traffic flow prediction method based on SVM, and the experimental results showed that the result of parallel SVM method is better than parallel BP neural network method, and when the number of parallel nodes is 100, the running time of two thousand links was 36.48 s [4]
Summary
Large-scale road network has high complexity, strong nonlinear, and high dynamic. The mass of traffic flow data brings enormous difficulties to traffic flow prediction. Li et al proposed a parallel traffic flow prediction method of space-time two-dimensional integration based on SVM, but this method is more suitable for emergency cases, not very practical for the normal traffic condition [1]. Wang et al implemented a parallel generalized neural network method for traffic flow prediction based on MPI (message passing interface) programming model. The experimental results showed that the speed of the proposed method was more than two times as fast as the serial method [3]. Wang proposed a parallel traffic flow prediction method based on SVM, and the experimental results showed that the result of parallel SVM method is better than parallel BP neural network method, and when the number of parallel nodes is 100, the running time of two thousand links was 36.48 s [4]. Its running efficiency was more than fourteen times as fast as the serial genetic neural network method [5]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
