The spread of fake news has led to an increasing worry about its negative impact on people. Thus, there is an emergence in research on fake news. Fake news detection on social media presents several challenges which make the existing data algorithms from traditional news media ineffective or not applicable. This research survey aims to present a comprehensive review of detecting fake news on social media which includes characteristics of fake news on psychology and social theories, existing algorithms from a data-mining perspective, evaluation metrics and representative data sets. The study also discusses related research areas, open problems and future research directions for fake news detection on social media.
Highlights:
- The study is divided into two phases of reviewing the existing literature on fake news: characterisation and detection. In the characterisation phase, the basic concepts and principles of fake news are introduced in both traditional and social media.
- In the detection phase, the study reviewed existing fake news detection approaches from a data mining perspective, including feature extraction and model construction.
- The research outlined four different directions that fake news research could take which were: data-oriented, feature-oriented, model-oriented and application-oriented.
- The research also proposes further research into fake news intervention with either 'proactive' or 'reactive' intervention methods.