Enhanced Stock Market Trend Prediction on the Indonesia Stock Exchange Using Improved Bacterial Foraging Optimization and Elitist Whale Optimization Algorithms

It was difficult to find low-risk firms in 2022 due to the large number of 825 listed on the Indonesia Stock Exchange (IDX). Essential concerns in financial works include stock price forecasting and price movement forecast. The stock market research has shown encouraging outcomes in the past. We still don't have a consensus on the finest stock value forecasting classical due to the fact that each study report has different architecture and uses different stock issuers. This paper presents a new method for optimizing bacterial foraging that uses optimal deep learning to forecast stock market trends. For the prediction process, the given model employs a multiplicative long short term memory (MLSTM) perfect, with IBFO serving as the model's fine-tuner. The research paper presents a modified whale optimization technique for feature selection. This research proposes an elitist whale optimization algorithm with the nonlinear parameter (EWOANP) to fix the issue where the whale optimization procedure does not strike a balance among exploration and exploitation and instead falls into the local optimum. To improve the odds of evading the local optimum, the shrinking encircling mechanism employs an elitist tactic based on random Cauchy mutation. Using the random Cauchy mutation as a starting point, the method cleverly selects the population with the best mutation solutions to move on to the next iteration. To strike a compromise among the two, the logarithmic spiral mechanism makes use of a nonlinear parameter. There are 2,588,451 entries in the dataset representing 727 IDX firms' open, highest, lowest, closed, and volume transactions. Various stock time records are used to examine the trial findings using current methodologies in terms of distinct metrics.