Improving the Method and Architecture of a Hybrid Machine Learning Model for Stock Price Prediction Using Text Classifiers

The aim of this research is to create a neural network for predicting stock prices, taking into account both internal data about the company and external factors such as news about the company. This project is important because of the growing interest and demand for artificial intelligence-based forecasting systems in the financial sector. Accurate and fast forecasts have a significant impact on investors' decisions. The main objective of the research is to develop an optimal model for predicting stock prices by analysing company data and news articles using machine learning techniques, in particular neural networks. The work will investigate the methods for estimating the tones of news articles and select the best method. The aim is to create a better model to predict stock prices more accurately based on available information.