Deep Learning-based Gated Recurrent Unit Approach to Stock Market Forecasting: An Analysis of Intel's Stock Data

Abstract

The stock price index prediction is a very challenging task that's because the market has a very complicated nonlinear movement system. This fluctuation is influenced by many different factors. Multiple examples demonstrate the suitability of Machine Learning (ML) models like Neural Network algorithms (NN) and Long Short-Term Memory (LSTM) for such time series predictions, as well as how frequently they produce satisfactory outcomes. However, relatively few studies have employed robust feature engineering sequence models to forecast future prices. In this paper, we propose a cutting-edge stock price prediction model based on a Deep Learning (DL) technique. We chose the stock data for Intel, the firm with one of the quickest growths in the past ten years. The experimental results demonstrate that, for predicting this particular stock time series, our suggested model outperforms the current Gated Recurrent Unit (GRU) model. Our prediction approach reduces inaccuracy by taking into account the random nature of data on a big scale.