A Novel Relevance Feedback Prediction Method (RFPM) for Predicting and Forecasting the Optimised OHLC Stock Price

After comparing and analysing a number of other neural network prediction techniques, time series models, and the challenges they provide in the context of stock price prediction, this study chose the autoregressive integrated moving average (ARIMA) and the long short-term memory (LSTM) neural networks. Finally, the methodologies' feasibility and model applicability are investigated by a detailed investigation of stock price prediction utilising the LSTM neural network enhanced by the RFPM algorithm. It has been shown that while developing future plans, investors significantly depend on historical data. Traditional financial market forecasting has mainly depended on opening and closing prices, but extreme highs and lows may also provide insight into the market’s future. As a result, the index of three representative stocks from the NSE is selected as the study object, and the starting price, closing price, lowest price, maximum price, date, and daily trading volume from these stocks are gathered. The findings demonstrate that the RFPM model can effectively predict the ideal future stock price. The basic concept is to predict stock price movements by diving into market history and determining the relevance of time series using the LSTM neural network model’s selective memory augmented deep learning function