A Comparative Analysis of Share Price Prediction and Trend Direction Using Sentiment Analysis of Financial News Articles

Stock price prediction is an area of current research due to the intricate data structure and many contributing elements. Many modern financial applications incorporate nonlinear and unpredictable components that exhibit temporal variability. The demand for solutions to highly nonlinear and time-dependent situations has experienced a significant increase in recent years. External variables such as public opinion and political events might impact the stock market. The main aim of this study is to examine the impact of investor mood on the market valuation of particular corporations. A machine learning model must consider emotional and contextual information in the first stages. The objective of this study is to analyze the influence of public opinion on the accuracy of algorithms in generating projections over 30 days. Furthermore, regression models are utilized to examine the interconnections across different organizations. The researchers obtained stock market data for their experiments from reliable sources, including the National Stock Exchange of India and the "Economic Times." The inquiry utilized the previously listed sources to gather monetary and public opinion statistics data. Both emotional and contextual elements are produced using pre-processing algorithms on the unprocessed textual input. Ultimately, the aforementioned characteristics are integrated into the data set creation process. A total of eleven discrete machine learning algorithms are employed to categorize the responses of news consumers. The interdependence analysis within the same industry unveiled a moderately positive correlation between stock markets. The present study examines the merits and demerits of artificial neural networks and hybrid intelligence, which are well-recognized artificial intelligence methodologies, within the context of the financial sector. In summary, the results affirm the integration of ensemble methodologies into the repertoire of algorithms employed in a pioneering investigation of predicting stock market fluctuations.