Sentiment Analysis of Stock Prices and News Headlines Using the MCDM Framework

Abstract

In the 21st century, the speedy progress in digital data procurement has led to the fast-growing amount of data kept in the database, data warehouses, or other data storehouses. The main reason behind this is that it provides a fast spreading of information and increases technology usage. The stock market is one of the utmost competitive financial markets where traders want to compute financial capacities with low latency and high output. In this study, we introduced an unsupervised MCDM-based Grey Relational analysis (GRA) model that targets giving appropriate sentiment tags to the news headlines and predicting the forthcoming stock prediction. To check the proposed model's applicability, we used INFOSYS and WIPRO datasets, which give satisfactory results over the proposed model. We recorded an accuracy of around 87%. We utilize a practical GRA model approach to evaluate and recommend the finest share stocks based on news headlines from multiple web sources. Our system's performance is evaluated using real-time data from WIPRO and INFOSYS.