Global News Sentiment Analysis

Document Type

Conference Proceeding

Publication Date

1-1-2021

Abstract

The Journalist’s Creed, a declaration of the principles, values, and standards of a journalist, states that a journalist should “believe that clear thinking and clear statement, accuracy and fairness are fundamental to good journalism”. However, in recent years there has been concern that personal, corporate, and government biases and opinions have had an effect on the fairness and integrity of modern journalism. Many mainstream media outlets are well known to have certain political inclinations and occurrences of “fake news” have become common. Studying mainstream international news sources is important for numerous reasons, such as national security, cultural stability, and the development of social/economic policy. Through the use of modern natural language processing techniques, we performed sentiment analysis on 10,809 articles published by ten globally distributed sources covering eleven globally impactful events, spanning a period of eleven years. We tested three open-source sentiment analysis tools and one commercial tool from Microsoft. We also tested three text processing approaches, i.e., sentence, paragraph, and article-based approaches. We observed that sentiment polarity was generally consistent across news agencies, however, the relative percent differences in sentiment magnitude varied by multiple standard deviations. Further, we found general agreement in sentiment polarity across the four analysis tools strengthening our finding regarding the news source content sentiment trends. The work provides both a reproducible computer-based metric for quantifying differences in global journalism and insights into the effective use of modern sentiment analysis tools.

Publication Title

Springer Proceedings in Complexity

First Page Number

121

Last Page Number

139

DOI

10.1007/978-3-030-77517-9_9

This document is currently not available here.

Share

COinS