Published Paper Details:

Stock market forecasts have always been the focus of experts in this and related fields. Over the years, many modern techniques have been used in conjunction with previously available statistical models to find better predictive techniques. Among modern technologies, machine learning and specifically artificial intelligence holds the greatest share of predictive models on the market. Compared to other techniques, deep learning techniques showed better results in modelling market movements. Some techniques have been tried and tested individually but with unsatisfactory results, these techniques include automatic feature extraction, time series prediction, recurrent neural network (RNN), etc. However, hybrid frameworks with multiple inputs and based on deep learning methods such as RNN and LSTM (Long Short-Term Memory) have not been studied much. We propose a recurrent neural network (RNN)-based framework that combines with long-short-term memory (LSTM) to predict the stock market closing price. The RNN-LSTM framework extracts features from a rich feature set and applies time series modelling to predict what will happen next. The set includes raw price data for the target index as well as foreign indices, technical indicators, exchange rates, and commodity price data, all selected by protocol and price configurations. known in the industry. Now let's come to the LSTM (Long Short-Term Memory) part, which is known for its series forecasting. LSTM is greatly helpful in time series forecasting. A correct price forecast will bring huge benefits to users of our framework/model. Successfully predicting the future price of a stock can bring substantial profits.