The authors have written an interesting book that summarizes problems related to detecting fake news.

In chapter 1, an introduction to the problem of fake news, the authors explain how fake news exploits individual gullibility: “(1) naïve realism--consumers tend to believe that their perceptions of reality are the only accurate views ... and (2) confirmation bias--consumers prefer to receive information that confirms their existing views” (p. 3).

Chapter 2, “What News Content Tells,” presents methods that attempt to determine whether news content is valid. It covers techniques that can be used to analyze the text of a suspect news article. A genuine news article is typically written in an inverted pyramid style: the most important information occurs early in the text, covering the familiar who, what, where, how, when, and why questions. A genuine news article will avoid emotionally charged words. This interesting chapter essentially analyzes the text of an article to determine whether it is genuine. The best method is manual fact checking, that is, a news researcher verifies the information using a number of sources. There are also websites where you can check some information.

Chapter 3 covers social context. It studies the user (or author) of an article. Information such as length of time affiliated with a certain website can be studied. It is assumed that somebody who joined a website a few days ago may be more likely to generate a fake news article than somebody who has belonged to a website for a number of years. The chapter also looks at propagation; for example, fake news can be passed on from one friend to another through social media networks.

In chapter 4, “Challenging Problems of Fake News Detection,” the topics covered include: 1) fake news early detection, 2) weakly supervised fake news detection, and 3) explainable fake news detection.

Appendix A covers data repositories. These include the Columbia Summarization Corpus, which extracts data from the academic project Columbia Newsblaster. Unfortunately, Newsblaster ended in 2016.

Appendix B covers various tools for tracking and detecting fake news on social media. Appendix C, “Relevant Activities,” includes educational programs, computational competitions, and research workshops and tutorials.

Overall, the authors cite a number of methods that can be used to detect fake news. The book is heavily referenced with more than 200 citations; there are also dozens of uniform resource locator (URL) references.

In summary, this brief, well-referenced book relies on dozens of equations. Readers can use it to become familiar with fake news detection methods. It can also be used as an advanced reference in classes attempting to study the detection of fake news.