A test platform for analysing retracted articles
RetractedIF is an experimental platform for generating analysis results on retracted articles and on open databases (Retraction Watch, SciELO, OpenAlex). It lets a closed research group view the results from different perspectives, to support studies and the development of more robust machine-learning and image-processing techniques for detecting problems in scientific articles. The date of every data source is shown on each result.
This is not a public service. The platform is closed and for testing: only members of the research group have access, and external users cannot sign up or use it. Results are experimental and are not a verdict on any article or author.
What it does
Article databases
Upload a list of DOIs, validate it on OpenAlex and keep it as a named database. Each DOI is stored once and shared by every database that lists it.
Retraction Watch
Query retractions by period, reason, country, subject, journal or publisher. Results come as charts, tables and CSV files, each stamped with the date of the Retraction Watch snapshot used.
SciELO × Retraction Watch
Retraction rates per collection, year and journal, with 95% confidence intervals. The reasons for retraction are compared with the whole Retraction Watch database.
Comparisons and reports
Compare databases by DOI to find overlaps, exclusive articles and their retraction status. Request full statistical reports developed by this research collaboration.
How long tasks work
Project and access
RetractedIF is part of a research collaboration between the Instituto Federal de São Paulo (IFSP), Câmpus Itapetininga and the Recod group. It was developed by Carlos Santos (IFSP).
Accounts are created by the administrator for members of the group only. External users cannot use the platform.
