Positive or Negative? A semantic orientation of financial news

Abstract

Semantic orientation, also known as sentiment analysis, is now expanding its research area due to its importance in many areas such as finance, business and management. In addition to the financial statements, news about the fundamentals of a company, forums, blogs and social media posts have become important sources affecting investors' decisions. On the other hand, due to the difficulties in monitoring the relevant and important stories about a company and determining its semantic orientation within this huge volume of information, automatic opinion mining has become a necessity for investors to act in a timely manner. In this context, this study presents a solution to the problem of semantic orientation of financial news by applying a Naive Bayes Classifier on a data set consisting of 75000 news texts formed from the news between 1996 and 2018 and analyzes the results in detail.