A Literature Review on Machine Learning Techniques and Strategies Applied to Stock Market Price Prediction

Abstract

Forecast of financial exchange price and patterns is considered as a significant challenging task and is of extraordinary consideration as anticipating stock costs effectively may prompt alluring benefits by settling on appropriate choices. Because of non-stationary, blasting, and noisy information, stock market price prediction is one of the major challenges, and the expectation among financial specialists is therefore difficult to contribute to profit making. This work of literature review presents the survey of more than 100 research papers proposing the various philosophies of machine learning methods as well as SVM classifier, Random forest, Neural Network, Bayesian model, Fuzzy classifier, Artificial Neural Networks, etc., in view of stock market price prediction. Among the cutting-edge techniques, various machine learning approaches and models due to their ability to identify complex patterns in a number of applications are studied. The review works are investigated utilizing certain dataset parameters, forecast technique used, and results achieved by various procedures. It can be observed that the stock market forecast is a very complicated activity. Various parameters should be considered to predict the market more accurately and effectively. This review shows that in financial market forecasting, machine learning algorithms continue to outperform most conventional stochastic methods.