Application of particle swarm optimization neural network in Financial Distress Analysis
Download as PDF
DOI: 10.23977/GEBM2020.030
Corresponding Author
Zhou You
ABSTRACT
At present, BP neural network has been successfully used in the financial analysis and prediction of the company, but the traditional gradient descent exploration method is adopted in the neural network, which causes the shortcomings of slow convergence speed and easy to fall into the local optimum, which has a bad impact on the application effect. In this paper, an improved PSO (particle swarm optimization) global search algorithm is proposed to train BP network. It not only retains the original advantages of neural network, but also overcomes the shortcomings of traditional training methods. The results show that this method is better than the traditional neural network model.
KEYWORDS
Financial distress, particle swarm optimization, neural network, training method