Prediction of Stock Market with Sentiment Analysis

• Produced by: PandasGo
• Team member: Lin Zeng(lz447), Qin(Ann) Lu(ql224), Sizhang(Cedric) Zhao(sz459)
• Abstract: In this project, we perform sentiment analysis and machine learning on publicly available U.S news (From Google News) to construct a measurement of "market sentiment". These estimated market sentiments and previous days' Dow Jones Industrial Average(DJIA) values are fed to our model learning framework which uses linear regression and Self-organizing Fuzzy Neural Network(SOFNN) to predict future DJIA movements. We also implement a portfolio management strategy to make appropriate buy/sell decisions based on our predicted values.