Exploratory analysis of text duplication in peer-review reveals peer-review fraud and paper mills

Research Integrity and Peer Review

This preprint sits comfortably within the existing paradigm of research integrity. It explores a specific anomaly within the peer-review process—text duplication—and investigates this anomaly using established methods and tools within data science. While the preprint identifies potential misconduct, it doesn't fundamentally challenge or seek to overturn current understanding of the peer-review process or research integrity itself. It contributes to the existing body of knowledge by providing further data and insights regarding the prevalence, detection, and analysis of fraudulent practices in peer review, all within the broadly accepted norms and standards of the field. Thus, it qualifies as an example of 'Normal Science.' There's a nuance, however, which tempts a secondary classification of 'Model Drift.' The increasing prevalence of paper mills and the tactics they utilize (as explored in this preprint) could be viewed as pushing the boundaries of the existing peer review 'model' and exposing limitations, perhaps a nascent stage of drift towards a need for more robust review models. However, this 'drift' aspect is not the dominant characteristic of the research, hence 'Normal Science' as the primary classification.