Analyzing and distinguishing fake and real news to mitigate the problem of disinformation
- PDF / 1,854,646 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 81 Downloads / 255 Views
Analyzing and distinguishing fake and real news to mitigate the problem of disinformation Alina Vereshchaka1 · Seth Cosimini2 · Wen Dong1
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Identifying fake news has become an important issue. Increasing usage of social media has led to an increase in the number of people who can be influenced, thus the spread of fake news can potentially impact important events. Fake news has become a major societal issue and a technical challenge for social media companies to identify and has led many to extreme measures, such as WhatsApp deleting two million of its users every month to prevent the spread of fake news. The current problem of fake news is rooted in the historical problem of disinformation, which is false information intentionally, and usually clandestinely, disseminated to manipulate public opinion or obfuscate the truth. Our work addresses the problem of identifying fake news by (i) detecting and analyzing fake news features (ii) identifying the textual and sociocultural characteristics fake news features. Keywords Fake news · Real news · Fake news identification · Data analysis · Deep learning · Sociocultural textual analysis
1 Introduction Identifying fake news has become an important issue both for the public and the academic communities. Increasing usage of social media has led to an increase in the number of people who can be influenced, thus the spread of fake news can potentially impact important events, such as elections or contribute to propaganda. * Alina Vereshchaka [email protected] Seth Cosimini [email protected] Wen Dong [email protected] 1
Department of Computer Science and Engineering, University at Buffalo, Buffalo, USA
2
Department of English, University of Nevada, Reno, Reno, USA
13
Vol.:(0123456789)
A. Vereshchaka et al.
Recent examples included the controversy created during the 2016 U.S. presidential campaign (Grinberg et al. 2019; Allcott and Gentzkow 2017) and the rise of global propaganda campaigns (Khaldarova and Pantti 2016; Rashkin et al. 2017; Bisgin et al. 2019) by influencing mainstream media and internet discussions. Fake news has become a major societal issue and a technical challenge for social media companies to identify and has led many to extreme measures, such as WhatsApp deleting two million of its users every month to prevent the spread of fake news. One of the other challenges for researchers is the lack of data, since most of the time this kind of dataset has to be manually labeled and there can be difficulties in keeping the consistency in these classification processes (Wu et al. 2019). Our research was conducted using the datasets extracted using the FakeNewsNet tool, and the results may not be generalizable to other datasets since they may contain different features and different formulations of misinformation. Therefore, there is a strong need for that tool that can early distinguishing fake news and help to stop the viral spread of such news. The current problem of fa
Data Loading...
