Stock Market Prediction with Historical Time Series Data and Sentimental Analysis of Social Media Data

In the Indian stock market, stock costs are viewed as exceptionally fluctuating due to various factors such as political decision results, bits of gossip, budgetary news, public safety events and so on. This fluctuation behavior makes it a difficult and challenging task to predict stock prices. The proposed work aims to develop a new methodology by combining sentiment analysis along with normal stock market prediction from time-series data using deep learning techniques. It extracts sentiments from news events, social media platforms particularly from twitter and incorporates the polarity of the sentiments to enhance the prediction accuracy. From the results and analysis, it can be observed that the proposed approach improved the forecasting accuracy after introducing the sentiment polarity scores. Hence, this model helps the investors to make a wiser and better decision about their investment.