A Novel Company Financial Risk Warning Method Based on BP Neural Network

This paper focuses on the problem of company financial risk warning, which is of great importance in modern company management. As BP neural network is a powerful tool to make state forecasting in complex system, in this paper, we propose a new company financial risk warning approach based on BP neural network. After demonstrating the main characterics of BP neural network, the proposed algorithm is given. The main innovations of this paper lie in two aspects. For the first aspect, we propose a hybrid model to make financial risk warning which combines BP neural network and particle swarm optimization together. For the second aspect, we select eighteen indicators from the five types of financial indicators to construct neurons of the BP neural network. Experiments demonstrate that the proposed approach can effectively company financial risk, and the financial risk estimated of our algorithm is very close to the experts´ evaluation.