An Early Risk Warning of Cross-Border E-Commerce Using BP Neural Network

In this paper, we propose an early warning model of credit risk for cross-border e-commerce. Our proposed model, i.e., KPCA-MPSO-BP, is constructed using kernel principal component analysis (KPCA), improved particle swarm optimization (IPSO), and BP neural network. Initially, use KPCA to reduce the index Next, inertia weight threshold network are searched MPSO. Finally, used training data 13 different enterprises e-commerce’s risk. To analyze efficiency our approach, five testing evaluation. The experimental results show that mean absolute error (MAE) root square (RMSE) lowest in comparison existing models have much better efficiency.