Sentiment Analysis and Stock Market Prediction-Using news to predict stock markets

Abstract

Abstract

The aim of this paper is to propose a relatively new approach regarding the influence of financial news on stock market prices. Previously, a lot of effort and research has been done to study the effect of historical stock market prices. This approach mainly involved time series numerical data to predict the future trend of stock markets. But, with the advent of superior computing power and computing techniques such as Natural LANGUAGE Processing, SVM etc., it becomes imperative to look beyond just numerical data. Thus, paper is one such effort to use textual data in the form of financial news to predict the market trends. The idea behind using textual data is that text inherently contains words that convey the sentiment of that text. The sentiment may either be positive or negative. Therefore, the idea is to analyse the sentiment of the text to classify the text as being positive or negative with the assumption that positive news will affect markets positively and negative news will affect markets negatively. However, textual data alone cannot be a good predictor of stock markets as there a number of other factors that influence stock markets. Hence, we do not focus on just textual data in this paper. We also take into account numerical aspects such as opening and closing prices of markets in confluence with the textual data to predict stock markets. This paper also presents a review of some of the most recent and most efficient research that has been done to find out the effect of public mood or sentiment on the stock markets. This paper also presents various findings of such researches.

Keywords: Stock markets, social media, sentiment analysis, machine learning

Cite this Article

Shah Abrar Amin, Dhajvir Singh Rai. Sentiment Analysis and Stock Market Prediction: Using News to Predict Stock Markets. Journal of Artificial Intelligence Research & Advances. 2019; 6(2): 17–24p.