MANIFESTO: a huMAN-centric explaInable approach for FakE news spreaders deTectiOn

Abstract

Fake news spreading is strongly connected with the human involvement as individuals tend to fall, adopt and circulate misinformation stories. Until recently, the role of human characteristics in fake news diffusion, in order to deeply understand and fight misinformation patterns, has not been explored to the full extent. This paper suggests a human-centric approach on detecting fake news spreading behavior by building an explainable fake-news-spreader classifier based on psychological and behavioral cues of individuals. Our model achieves promising classification results while offering explanations of human motives and features behind fake news spreading behavior. Moreover, to the best of our knowledge, this is the first study that aims at providing a fully explainable setup that evaluates fake news spreading based on users credibility applied to public discussions aiming to a comprehensive way to combat fake news through human involvement.

Legal Strategies for Combating Fake News and Misinformation

“…Digital platforms play a pivotal role in the dissemination of misinformation, a theme consistently highlighted across studies. Lampridis, Karanatsiou, and Vakali (2022) propose a human-centric explainable approach to detecting fake news spreaders, emphasizing the importance of transparency and accountability in digital platforms' operations (Lampridis et al, 2022). This aligns with our findings on the critical role of digital platforms in content moderation and the need for collaboration with legal authorities to curb misinformation effectively.…”

“…The role of digital platforms in this context is pivotal, as they serve as the primary medium through which misinformation is disseminated. Lampridis, Karanatsiou, and Vakali (2022) advocate for a human-centric explainable approach to detecting fake news spreaders, emphasizing the need for transparency and accountability in digital platforms' operations (Lampridis et al, 2022). Legal responses to fake news have varied globally, with some countries implementing stringent measures to curb misinformation while others focus on selfregulation by digital platforms.…”

A System to Ensure Information Trustworthiness in Artificial Intelligence Enhanced Higher Education

“…Current techniques for detecting fake information do not integrate both human and technological knowledge (Lampridis et al, 2022). A human-centered information assessment detection system is important for several reasons.…”

“…For example, the trustworthiness of a text is strongly associated with such properties as information content, neutrality, clarity, sentiment, and logic (Kauttonen et al, 2020). Second, a human-centered approach helps ensure ethical considerations are considered in the detection and assessment of fake information (Lampridis et al, 2022). Machines can make decisions based purely on data without considering the broader ethical implications.…”

“…Incorporating both human and technological knowledge (Lampridis et al, 2022) makes EMIAS a robust solution that leverages humans' ability to interpret complex and contextual information. The integration of transparency and explainability features into its operation ensures users have clear insight into its decision-making processes, fostering trust and promoting wider adoption, the issues mentioned by Scharowski et al (2023) and Szczepański et al (2021).…”