Stock Market Prediction Based on Big Data Using Deep Reinforcement Long Short-Term Memory Model

Abstract

In this modern era, the stock market is one of the important platforms for several business developments to ensure their growth for upcoming years. Big data is another technical aspect of the stock market to import large amounts of stocks. Deep Reinforcement Long Short Term Memory (DRLSTM) model is proposed for achieving better prediction rate for stock market trends based on the technical indicators. Three most popular banking organizations data is obtained in real-time live stocks from the NIFTY-50 market data. The data is enclosed based on trading days from 2000 to 2020. The bid data approach known to be a Hadoop framework is used simultaneously to handle large amounts of data for processing through distributed storage. The experimental results are performed based on the mean squared error (MSE) for the proposed model which obtained a low error rate of 0.017% for SBI, 0.014% for HDFC, and 0.018% for BajajFin. The proposed model is evaluated and results are compared with other existing techniques which the RLSTM outperforms by obtaining a high accuracy rate.