Evidence-Aware Fake News Detection: A Review

The proliferation of fake news on social media platforms has raised concerns about the reliability and trustworthiness of online information. To combat this issue, automatic fake news detection systems have been developed, leveraging advanced techniques such as machine learning and natural language processing. However, these systems face challenges in accurately discerning genuine and fabricated news. In recent years, there has been a growing recognition of the importance of integrating evidence and external knowledge into these detection models. External knowledge, including fact-checking databases, trusted reference sources, and credible news outlets, provides valuable context and verification to assess the credibility and accuracy of news articles apart from the available content. By cross-referencing claims with external knowledge, automatic detection models can identify potential instances of fake news, verify claims, assess source credibility, and enhance their overall effectiveness. This paper provides a taxonomy of evidence-aware fake news detection. This paper further explores the various ways in which external evidence can be exploited for automatic fake news detection, including fact-checking partnerships, metadata analysis, social context analysis, and user-generated content analysis. We also propose a multimodal evidence-aware FND framework and further examine the benefits and challenges associated with integrating external knowledge and evidence into detection models, and future possibilities contributing to the development of more trustworthy and informed digital environments.