Artificial Neural Network-Based Fuzzy Interference System for Stock Market Price Prediction for Profit Maximization

Regarding stocks, the stock market's lack of clearness and complexity is experiencing a state of not clear, and hard to find the non-linear relations that are within the data. As a result, the need for sophisticated modeling techniques to identify the stock market patterns is underlined. In this paper, we use the Adaptive Neuro Fuzzy Inference System (ANFIS), a soft computing model hybrid that integrates fuzzy logic with artificial neural networks (ANN), an efficient technology in market forecasting accuracy and reliability under dynamic conditions. There is a historical data set of Samsung stock prices that are subjected to input labels, such as same-day, opening price, highest price, lowest price, closing price, and adjusted close price, to forecast the opening price of the next day. The various membership functions, which include Gaussian, trapezoidal, and triangular, are incorporated into this model for comparative analysis, and the best fit is analyzed.