DEEP LEARNING-BASED STOCK PRICE PREDICTION SYSTEM AND METHOD USING RECURRENT NEURAL NETWORK

摘要

The present invention relates to a deep learning-based stock price prediction system and method using a recurrent neural network, wherein the system can improve the accuracy of prediction by machine learning various past time series data related to a stock through a deep learning model using a recurrent neural network in predicting a stock price at a next time point relative to a reference time point of a time series, then predicting the stock price with a value that represents a fluctuation of the stock price at the next time point relative to the stock price at the reference time point as a percentage or a value corresponding to the fluctuation, and dividing the prediction into high and low prices of the next time point period relative to the reference time point, and can utilize the prediction results in real transactions such as stocks or related derivatives and funds, and also enhance a rate of return on such investments.