Particle swarm optimization based BP neural network on credit risk evaluating control system

The purpose of this paper is to optimize evaluating control system, describe the topology structure of particle swarm optimization (PSO) based BP neural network (PSOBPNN), introduce the principle of PSOBPNN, and give the implement step of PSOBPNN. The algorithm is applied to electronic commerce enterprise credit risk evaluating system, then compare its result with conventional BP neural network. The comparing result shows that PSOBPNN fits to complex system such as credit evaluating control system, it improves in a certain extent on training speed and precision, it can improve the quality of electronic business enterprise credit evaluating control system, and it fits to solve some problems in which evaluating indexes weights are difficult to be determined or there exists complex non-linear relation among them.