Tool for automatically detecting predatory journals and conferences using multiple authoritative data sources
-
Updated
Sep 14, 2026 - Python
Tool for automatically detecting predatory journals and conferences using multiple authoritative data sources
Predatory Conferences & Events — Open Dataset
A Firefox and Google Chrome extension to detect whether the journal you're viewing is a predatory journal
Reference-integrity checks for papers: DOI validation, retractions, expressions of concern, preprints, OA, predatory journals.
Explainable AI platform for discovering conferences, journals, CFPs and research positions, with hybrid semantic search, venue risk scoring and personalized recommendations.
Maurice Zeegers / Meta Research
To associate your repository with the predatory-journals topic, visit your repo's landing page and select "manage topics."