An In-Depth Review of Machine Learning Techniques for Multimodal Stock Market Prediction from an Empirical Perspective

Predicting the stock market is a crucial but difficult task due to the complexity and volatility of financial markets. In recent years, multimodal techniques utilizing the fusion of diverse data sources have demonstrated great promise for enhancing the accuracy of stock market prediction models. From an empirical standpoint, this paper provides a comprehensive review of machine learning techniques for multimodal stock market prediction. This review's primary objective is to compare and assess the functionality, precision, accuracy, recall, delay, and scalability of various multimodal techniques. By analyzing a large number of empirical studies, researchers shed light on the strengths and weaknesses of various multimodal approaches used for stock market forecasting. In addition, researchers develop an effective Stock Market Predictive Rank (SMPR) to address the need for a comprehensive evaluation metric that combines multiple performance measures. SMPR evaluates the capability of a model to handle multimodal data, in addition to its precision, accuracy, recall, delay, and scalability. By assigning a rank based on these metrics, researchers hope to identify the best models for predicting stock prices in real-time situations. This review is necessitated by the growing interest in multimodal techniques for stock market forecasting and the absence of a comprehensive analysis of various performance metrics. By synthesizing the existing literature, researchers provide a holistic perspective on the advantages and disadvantages of various multimodal approaches, thereby facilitating future research and practical application. researchers aim to provide researchers, practitioners, and decision-makers with valuable insights into the current state of multimodal stock market prediction techniques through this review. The findings presented in this paper can aid in the selection and development of effective models for real-time stock market forecasting, thereby enhancing investment decision-making and financial risk management process