Research on the Prediction of Artificial Intelligence Enterprise Value Based on Improved PSO and BP Neural Network

An important prerequisite for promoting the healthy development of the artificial intelligence industry is to make a more scientific and accurate prediction of the value of artificial intelligence enterprises, so as to better realize the effective allocation of resources in the investment market. In view of the wide application of BP neural network in the prediction of nonlinear fields, this paper uses particle swarm optimization algorithm and BP neural network algorithm to construct an artificial intelligence enterprise value prediction model based on improved particle swarm optimization algorithm and BP neural network, and uses Matlab tools to analyze and predict the value evaluation indicators of artificial intelligence enterprises. This study selects the daily stock trading data of IFLYTEK, a leading enterprise in the artificial intelligence industry, from 2020 to 2023. The results showed that the RMSE, MAE and PR of BP neural network were 1.275, 0.724 and 98.690%, and the RMSE, MAE and PR of PSO-BP neural network were 0.586, 0.389 and 99.246%. It shows that compared with BP neural network prediction model, the relative error of actual data of PSO-BP prediction model is smaller and the accuracy is higher, which can provide an effective model for the subsequent artificial intelligence enterprise value evaluation.