Abstract
In recent years, the development of high-speed railway industry in China is very rapid. However, the development of Chinese high speed railway cannot be further improved without basic research. The passenger flow is the basis and foundation to build high-speed railway. Therefore, to establish a set of analysis method for big data forecasting of railway passenger flow has great theoretical value and practical significance. In this paper, a parallel particle swarm optimization algorithm which is based on Hadoop framework is proposed, which can effectively avoid the particle swarm algorithm falling into local extreme value. Parallel particle swarm optimization algorithm is used to optimize the parameters ( C,??2 ) of SVM.
Taking into account the each solution of particle swarm adaptation value is to go through the quadratic optimization process of support vector machine, we use the particle swarm optimization in parallel computing to complete the rapid prediction of big data. Experimental results show that the algorithm has good performance and high accuracy, which proves the validity of the algorithm.
