Semester of Graduation
Information Systems and Business Analytics
First Major Professor
Master of Science (MS)
We live in an age where machine learning and data science, in general, are influencing our decision-making capabilities in all aspects of life. We now depend heavily on historical data to make crucial decisions about our health, whether we want to go ahead with surgery or choose alternate paths. One field where the influence of such high-end cutting-edge technologies plays a crucial part in the financial sector, whether it is the use of Reddit pages to create a significant shift away from large financial corporations or algorithmic trading to enhance the profitability of our investment portfolio, the stock market is one of the most considerable investment hubs for everyone. It is influenced by the way a company performs and the sentiment of the people towards the company and its products. One significant analysis that can be performed to predict the stock values better is the news-based sentiment analysis of a stock. The methodology is to get data from various news castors using an Application Programming Interface (API), clean the data and perform complete sentiment analysis to understand the correlation between stock value and news. Using models like Linear Discriminant Analysis (LDA), Linear Regression, etc., we will try to find out whether we can predict a stock's value increase or decrease in the future. The model evaluation metrics used are Training and Testing accuracies, Precision, Recall, and Confusion matrix.
Mylavarapu Vishnu Kalyan
Embargo Period (admin only)
Mylavarapu, Vishnu Kalyan, "News based prediction of Stock price" (2021). Creative Components. 871.
Artificial Intelligence and Robotics Commons, Business Analytics Commons, Business Intelligence Commons, Databases and Information Systems Commons, Data Science Commons, Programming Languages and Compilers Commons