Determining Stock Market Prediction Using Opinion Mining

Abstract

The field of stocks or securities exchange has faced many transformations over the years. While the only thing constant in the stock market domain is its volatility, it still remains the most powerful investment strategy if you want to make money even when you sleep. In this paper, we propose an online website , that will make predictions on the performance of stocks in the future and help investors reach the most accurate decisions about investments. For this, we use a machine learning model, the base of which is a linear regression algorithm. Recent studies have suggested using a support vector machine (SVM) to make the predictions, which lead to a look-ahead bias, which was then tackled by coupling it with a rolling window approach to improve its accuracy. We, on the other hand, use linear regression to make predictions. The results of our experiment show that our method is more accurate as compared to SVM. Furthermore, any new investor who needs to brush up his knowledge about stocks will be redirected to articles or teaching videos about what the stock market is.