The study of SVM optimized by Culture Particle Swarm Optimization on predicting financial distress

In this paper, we applied culture particle swarm optimization algorithm (CPSO) to optimize the parameters of SVM. Utilizing the colony aptitude of particle swarm and the ability of conserving the evolving knowledge of the culture algorithm, this CPSO algorithm constructed the population space based on particle swarm and the knowledge space. The two spaces evolved independently, at the same time, the population space continuously transferred the evolving knowledge to the knowledge space, and then the knowledge space to achieve global optimization. Additionally, the proposed CPSO-SVM model was test on the prediction of financial distress of listed companies in China. Then we compared the accuracies of CPSO-SVM with other models (Standard SVM, PSO-SVM and PSO-BPN). Experimental results showed that CPSO-SVM performed the best prediction accuracy and generalization.