The stock market has the characteristics of large fluctuations and high dimensions, and can be regarded as a nonlinear time series system, so the traditional time series method is not applicable. For the study of deep learning methods, this paper proposed a bidirectional long-short term memory (BLSTM) neural network. Compared with ARIMA model and LSTM neural network, the (BLSTM) neural network is used to predict the accuracy of GREE stock price. Firstly, the stock data is normalized and pre-processing, and then the processed data is input into one-way and twoway LSTM respectively. In the neural network, Dropout is used as an optimization term to prevent over-fitting of the network, and then the three evaluation criteria of RMSE, MAE and Loss are selected to comprehensively analyze the rationality of a single bidirectional LSTM neural network. The experimental results show that the RMSE and MAE are reduced by about 24.2% and 19.4% respectively, and the deviation accuracy is increased by 0.13%. The network complexity is not high, and the error is effectively reduced. It can provide reference value for short-term market investors.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.