Fake News Detection Regarding the Hong Kong Events from Tweets

Abstract

The rapid development of network services has led to the exponential growth of online information and the increasing number of social media users. These services are exploited by malicious accounts that spread fake news and propaganda in vast user networks. Consequently, an automated solution for fake news and deception detection is required. This paper introduces a new data set consisting of 2,366 tweets written in English, regarding the Hong Kong events (August, 2019), and a well-defined method for fake news detection that uses both linguistic and network features. Our approach is tested with experiments using 2 machine learning models, achieving high performance compared to previous research.