News Analytics and Sentiment Analysis to Predict Stock Price Trends

Business news carries varied information of different companies. But in this rapidly moving world the number of news sources present is uncountable, and it’s humanly impossible to read and find all relevant information in the form of news to draw a conclusion timely to make an investment plan that returns maximum profit. In this paper, we have proposed a predictive model to predict sentiment around stock price. First the relevant real time news headlines and press-releases have been filtered from the large set of business news sources, and then they have been analyzed to predict the sentiment around companies. In order to find correlation between sentiment predicted from news and original stock price and to test efficient market hypothesis, we plot the sentiments of 15 odd companies over a period of 4 weeks. Our result shows an average accuracy score for identifying correct sentiment of around 70.1%. We also have plotted the errors of prediction for different companies which have brought out the RMSE and MAE of 30.3% and 30.04% respectively and an enhanced F1 factor of 78.1%. The comparison between positive sentiment curve and stock price trends reveals 67% co-relation between them, which indicates towards existence of a semi-strong to strong efficient market hypothesis. Keywords— stock price trends, prediction model, knowledge discovery, sentiment analysis, market trends, news analytics, efficient market hypothesis.