LSTM Approach for Efficient Stock Market Prediction

Abstract

Stock market investing has always been difficult for shareholders and prevents the use of standard models to make more accurate predictions of future values. Although many researchers and academicians have proposed methods to make stock price prediction more efficient. But after going through those proposals, we found a number of loopholes that can be tackled using a different approach. In this research work, machine learning and a study of finance have been combined to construct a model employing long-short term memory (LSTM) that forecasts the value of the SENSEX in the future. Finally, we have evaluated the performance of our proposed method. So this research work can be used by other researchers in the same domain. Our research will encourage practitioners to better identify the exciting sector for future views while also assisting beginners in comprehending the ML paradigm.