A Survey on Natural Language Processing for Fake News Detection

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 6086–6093 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 6086 A Survey on Natural Language Processing for Fake News Detection Ray Oshikawa1, Jing Qian2, William Yang Wang2 1College of Arts and Sciences, The University of Tokyo 2Department of Computer Science, University of California, Santa Barbara Tokyo, 153-8902 JAPAN / Santa Barbara, CA 93106 USA ray.oshikawa@gmail.com, {jing qian,william}@cs.ucsb.edu Abstract Fake news detection is a critical yet challenging problem in Natural Language Processing (NLP). The rapid rise of social networking platforms has not only yielded a vast increase in information accessibility but has also accelerated the spread of fake news. Thus, the effect of fake news has been growing, sometimes extending to the offline world and threatening public safety. Given the massive amount of Web content, automatic fake news detection is a practical NLP problem useful to all online content providers, in order to reduce the human time and effort to detect and prevent the spread of fake news. In this paper, we describe the challenges involved in fake news detection and also describe related tasks. We systematically review and compare the task formulations, datasets and NLP solutions that have been developed for this task, and also discuss the potentials and limitations of them. Based on our insights, we outline promising research directions, including more fine-grained, detailed, fair, and practical detection models. We also highlight the difference between fake news detection and other related tasks, and the importance of NLP solutions for fake news detection. Keywords: Natural Language Processing, fake news detection, survey. 1. Introduction Automated fake news detection is the task of assessing the truthfulness of claims in news. This is a new but critical NLP problem because both traditional news me- dia and social media have huge social-political impacts on every individual in the society. For example, expo- sure to fake news can cause attitudes of inefficacy, alien- ation, and cynicism toward certain political candidates (Balmas, 2014). Fake news even relates to real-world vi- olent events that threaten public safety (e.g., the PizzaGate (Kang and Goldman, 2016)). Detecting fake news is an im- portant application in the world that NLP can help with, as it also creates broader impacts on how technologies can fa- cilitate the verification of the veracity of claims while edu- cating the general public. The conventional solution to this task is to ask professionals such as journalists to check claims against evidence based on previously spoken or written facts. However, it is time- consuming and expensive. For example, PolitiFact1 takes three editors to judge whether a piece of news is real or not. As the Internet community and the speed of the spread of information are growing rapidly, automated fake news de- tection on internet content has gained interest in the Ar- tificial Intelligence research community. The goal of au- tomatic fake news detection is to reduce the human time and effort to detect fake news and help …

Content truncated to 500 words