Stock Price Prediction Using Concept of Long-Short Term Memory

Devanshu Upadhyay, Soumya Sharma

Abstract: Stock Market is a source of investment for nearly every person. People invest their complete earnings and savings to make easy money. They are not aware of the spontaneous nature of the market. This nature of the market can lead to people making huge profits or losing everything. This uncertainty in the market may lead to people losing faith while investing their money in stock market. In order to make people aware of the stock market?s nature and providing them knowledge on how to invest, this paper proposes a model built on the concept of Long-Short Term Memory. It is a special type of recurrent neural network that can remember the important past sequences of time-series data. The model will predict the future stock prices and will provide a comparative report to determine the model?s accuracy and for better analysis, while investing in stocks.

Keywords: Stock Market, Stock Prices, Prediction, Machine Learning, Long-Short Term Memory, Recurrent Neural Network