I Models for Misinformation Detection in Social Media: A Survey of Techniques, Applications, and Challenges

Abstract

n the era of ubiquitous digital communication, social media platforms have become primary channels for information dissemination. However, this unprecedented connectivity has also facilitated the rapid spread of misinformation, posing significant risks to public safety, democracy, and social cohesion. Detecting and mitigating misinformation is a critical challenge for governments, media platforms, and technology providers alike. In recent years, artificial intelligence (AI) models, including machine learning (ML) and deep learning (DL) techniques, have shown promising capabilities in automatically identifying and flagging false or misleading content. This survey paper systematically reviews the current state of AI-driven misinformation detection on social media, examining conventional machine learning models, state-of-the-art deep learning architectures, hybrid frameworks, and ensemble techniques. The study also explores applications, publicly available datasets, evaluation metrics, and highlights key challenges such as data imbalance, context dependency, and multilingual misinformation. Finally, the paper proposes future research directions emphasizing explainability, cross-platform detection, multimodal misinformation identification, and ethical considerations in AI deployment for content moderation