Research on Computer Prediction Model Using SSA-BP Neural Network and Sparrow Search Algorithm

Abstract

Stock price forecasting has always been a very challenging topic and a hot gampit in the wealth management and its applications. Timely and accurate forecasts can not only provide individuals with excellent investment opportunities, but also help the country's macroeconomic regulation and control. In order to achieve accurate predictions, people have tried various methods, among which machine learning methods have made breakthroughs. In this article, we propose a BP neural network model optimized based on the Sparrow Search Algorithm (SSA), and we use it to forecast the stock market. The experimental results show that the SSA-BP neural network model effectively overcomes the problem that the BPNN and the PSO-BPNN are easy to fall into the local optimum. The accuracy is better than BPNN, PSO-BPNN and LSTM model.