Finance Market Prediction: A Hybrid Analysis Model

Abstract

We introduce a novel approach to predict the future of Finance market by applying Machine Learning principles and Sentiment Analysis. In this approach we use the combination of Sentiment Analysis of twitter data and Machine learning approach for historical data, to predict the stock market stability for the next day. Stock prices are considered to be very dynamic and movable to quick changes because of the underlying nature of the stock market and in part because of the mix of known parameters (previous day opening price, previous day closing price etc.) and unknown factors (like Rumors, election etc.). Using machine learning approach we use known parameters for calculating the next day status of finance market and by using sentiment analysis of twitter data we evaluate the finance market status for next day for unknown factors. In the combination of both approach we calculate the stock market value for next day.
Keywords: Naïve Bayes Classifier, Sentiment Analysis, Sentiment Classification, Twitter.