Stock Price Prediction System with Improved LSTM

High Volatility in the Stock Market makes forecasting the stock price challenging globally. With the introduction of state-of-the-art technologies like Artificial Intelligence and Machine Learning, programmed prediction methods have proven to be more significant in forecasting the prices of stocks. In this work, the Recurrent Neural Network-based model, LSTM, is utilized to forecast the day's high and low costs of the three tyre companies: MRF Limited, CEAT Limited, and JK Tyre Industries and Limited, based on the company's historical data. The LSTM model uses technical values like RSI, SMA, etc., for training to make better predictions with higher accuracy. The system uses standard indicators: MAPE and MSE, Precision-Recall, etc., to evaluate the model's performance.