Analysis of Tweets for Prediction of Indian Stock Markets

The information age has greatly contributed to the massive data (big data) available on the web today. The proliferation of social media such as Twitter and Facebook has also contributed to the large volumes of data. Sense can be brought to this data by finding patterns in the data thereby extracting valuable information that can help businesses stay ahead of competition. People tend to show their emotions on Twitter therefore it is possible to mine behavioral data from these platforms. The stock market is one such area where behavioral data can be used to determine stock movement. The hypothesis therefore is that there should be a correlation between trends on the stock market and the emotions exhibited by people on social media platforms. This paper focusses on prediction of daily stock movements of three Indian companies listed on National Stock Exchange (NSE). The Support Vector Machine (SVM was used for prediction and its performance was evaluated on real market data. Experimental results showed stock movement predictions that were above the baseline. The results were very promising but contained random variations across different datasets.