Sentiment Analysis for Financial Predictions: A Review

Siddharth Chaurasia

Abstract

AbstractSentiment analysis is one of the popular research areas in computer science, as it finds its application in numerous areas like finance, election and product reviews. Behavioral economics suggests that decision of investing in financial markets is driven by feelings like greed, fear etc. With widespread adoption of technology and digitization of corporate disclosures, news and social media platforms, and information expressed through words can be mined and used to gaze into future. Text streams are inherently challenging to interpret the meaning and its effects. Current paper reviews the recent work done for financial market predictions by processing texts from varied sources using machine learning techniques. Support vector machine and Naïve Bayes turn out to be extensively used for financial predictions by analyzing text data. Moreover, almost all the studies that used disparate sources of information showed improvement when sentiments were included for predictions, thus highlighting the importance for its inclusion in the prediction process.Keywords: Machine learning, prediction, sentiment analysis, stock market, text miningCite this ArticlePuneet Misra, Siddharth Chaurasia. Sentiment Analysis for Financial Predictions: A Review. Journal of Artificial Intelligence Research & Advances. 2019; 6(1): 59–68p.