Stock Market Prediction and Risk Analysis using NLP and Machine Learning 
Nishant Verma1, S G David Raj2, Ackley J Lyimo3, Kakelli Anil Kumar4
1Nishant Verma*, PG Scholar, School of Computer Science Engineering Specialization in Information Security, Vellore Institute of Engineering and Technology, Vellore, Tamil Nadu, India.
2S G David Raj, PG Scholar, School of Computer Science Engineering Specialization in Information Security, Vellore Institute of Engineering and Technology, Vellore, Tamil Nadu, India.
3Ackley Lyimo, PG Scholar, School of Computer Science Engineering Specialization in Information Security, Vellore Institute of Engineering and Technology, Vellore, Tamil Nadu, India.
4Dr. Kakelli Anil Kumar, Associate professor, School of Computer Science Engineering specialization in Network Security, Vellore Institute of Engineering and Technology, Vellore, Tamil Nadu, India.
Manuscript received on May 30, 2020. | Revised Manuscript received on June 08, 2020. | Manuscript published on June 30, 2020. | PP: 813-815 | Volume-9 Issue-5, June 2020. | Retrieval Number: D6648048419/2020©BEIESP | DOI: 10.35940/ijeat.D6648.069520
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The stock market has been an instrument of investment for more than 200 years now. The price movements in the stock market have been an enigma for many financial analysts and they have tackled this problem with very little success. The phenomenal advancement in technology led to increased storage systems, higher processing speed and better algorithms. Thus, it is more possible now to develop a system for predicting stock markets. People have this taboo that only big investors can profit from the stock market and it is a trap for retail investors or small players. The solution we propose here is easy to understand and implement. We first do the sentiment analysis of the selected stock and then suggest whether to buy, sell or hold. Secondly, we calculate the maximum risk involved in the investment using a threefold approach; market risk, sector risk and stock risk. Finally, using Support vector Regression (SVR) with three different approaches, we calculate expected return and compare them with actual returns.
Keywords: Machine Learning, Risk Analysis, Stock Market, Stock Prediction, Support Vector Regression (SVR).