Survey and review papers¶
Survey and review papers provide comprehensive syntheses of research areas, establishing the state of knowledge, identifying gaps, and setting research agendas. These foundational works are essential for researchers entering a field or seeking an authoritative overview.
Key observations¶
Surveys establish baseline knowledge:
High-quality reviews systematize existing findings, identify contradictions, and reveal where evidence is lacking—critical for directing future research efforts.
Multidisciplinary surveys bridge silos:
Misinformation research spans psychology, computer science, political science, communication, law, and security. Cross-disciplinary surveys break down communication barriers and foster methodological innovation.
Reviews surface unanswered questions:
A well-structured review highlights not just what is known but what remains unknown, making explicit the frontier of the field.
Key papers in this wiki¶
- Advancing Graph Representation Learning with Large Language Models: A Comprehensive Survey of Techniques — Comprehensive survey of integrating Large Language Models with Graph Representation Learning; proposes novel taxonomy decomposing models into primary components (knowledge extractors and organizers) and operation techniques (integration and training strategies); identifies future research directions in graph foundation models.
- Shu et al. (2020) — Mining Disinformation and Fake News: Concepts, Methods, and Recent Advancements — comprehensive book chapter covering three pillars: (1) user engagement in information disorder dissemination; (2) detection and mitigation techniques using weak social supervision; (3) trending issues (fake news literacy, neural generation, blockchain); introduces weak social supervision paradigm and models (TriFN, dEFEND, MWSS) leveraging social media signals as training labels.
- Rana et al. (2022) — Deepfake Detection: A Systematic Literature Review — systematic review of 112 deepfake detection studies (2018–2020) with rigorous SLR methodology; categorizes techniques into four families (deep learning 77%, machine learning 18%, statistical 3%, blockchain 2%); compares detection approaches across datasets and metrics; identifies standardization gaps and future research priorities.
- Shu et al. (2017) — Fake News Detection on Social Media: A Data Mining Perspective — foundational survey organizing fake news through characterization (psychological and social theories) and detection methods (content and social context approaches); reviews datasets, evaluation metrics, and identifies future research directions.
- Sharma et al. (2018) — Combating Fake News: A Survey on Identification and Mitigation Techniques — comprehensive survey organizing detection methods into content-based (linguistic, deep learning) and feedback-based (propagation patterns, temporal dynamics, user engagement); systematic enumeration of 23+ datasets with detailed characteristics; proposes research directions in dynamic knowledge bases and intervention strategies.
- Lazer et al. (2018) — The Science of Fake News — multidisciplinary Science article synthesizing knowledge on fake news prevalence, impact, psychological mechanisms, and interventions (individual and platform-based); identifies major gaps and calls for industry-academic collaboration.
- Zhou & Zafarani (2020) — A Survey of Fake News: Fundamental Theories, Detection Methods, and Opportunities — comprehensive survey of fake news detection methods from linguistic, network, and stylometric approaches; categorizes datasets and benchmarks.
Related concepts¶
- Misinformation interventions — strategies to reduce spread and impact
- Fake news detection — methods to identify false content
- Cognitive mechanisms of misinformation — why people believe false claims