Key points are not available for this paper at this time.
Abstract Combating fake news is a crucial endeavor, yet the complexity of the task requires multifaceted approaches that transcend singular technological solutions. Traditional fact-checking, often centralized and human-dependent, faces scalability and bias challenges. This paper introduces a novel blockchain-based framework that leverages the wisdom of the crowd for an authority-free, scalable, automated and reputation-driven fact-checking. Within this framework, stance detection acts as an automated means of opinion retrieval, while the Proof of Reputation consensus mechanism fosters an environment where reputable contributors have greater influence in shaping news credibility. Concurrently, the Hoeffding bound is used to allow the system to adapt to evolving contexts. In contrast to Machine Learning -based approaches, our framework avoids the need for periodic retraining to update a model’s frozen knowledge of the world. The experimental study conducted on real-world data demonstrates that the proposed framework offers a promising and efficient solution to combat the spread of fake news.
Salah et al. (Wed,) studied this question.