Research On Stock Prediction And Application Based On LSTM Siamese Network Model

The development of the stock market is of great significance to the market economy.Many researchers also predict the trend of the stock market.Through some techniques to predict the future development trend of the stock and predict its subsequent rise and fall.Stock index is an important index to predict the economy.Stock index represents the overall development level and trend of the stock market.However,there are many factors affecting the fluctuation of stock index.With the establishment of financial market,the analysis and prediction of stock price is becoming more and more difficult.This paper selects the daily historical data of Shanghai and Shenzhen stock index,and selects four attributes:opening price,highest price,lowest price and closing price,of which the first three are used as independent variables to predict the rise and fall percentage of the next day.The twin network model of long-term and short-term memory is used to predict the stock index,and the results are compared and analyzed.The experimental analysis shows that this method has a good effect on the prediction of stock market trend.Finally,the algorithm proposed in this paper is applied to the stock trading system to give scientific indicators for users’ investment decision-making and improve the efficiency of stock trading.After the stock prediction model is built,the stock prediction model is applied to the construction of information system.This paper first collects the relevant data of the subject,analyzes the existing research,and then obtains the requirements of the stock trading management system realized in this paper from the analysis results.The system is analyzed from the functional and non functional requirements of the system,and complete the system design on this basis.