Pre-processing of Retracted Scientific Articles Using Machine Learning

• Grant number: 26/11981-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2026
• End date: August 31, 2027
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Carlos Henrique da Silva Santos
• Grantee: Mariana Alice Pires Leite
• Host Institution: Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP). Campus Itapetininga. Itapetininga , SP, Brazil
• Associated research grant: 23/12865-8 - Horus: artificial intelligence techniques to detect and forestall synthetic realities, AP.TEM

Abstract

The number of retracted articles in scientific literature has been increasing, a phenomenon primarily driven by two factors. First, the widespread adoption of Artificial Intelligence technologies, such as Generative Pre-trained Transformer (GPT) models, which facilitate the production of synthetically generated content that closely mimics human writing. Second, ongoing challenges to scientific integrity, ranging from conventional plagiarism to the rise of "paper mills"' and the manipulation of data and evidence. These issues undermine the reliability of the academic record and pose significant risks to science, particularly in sensitive fields such as health and biomedical sciences. In this context, this project proposes to investigate the prevalence of retractions in the biomedical literature, build an updated dataset containing retracted articles, and develop a procedure for extracting structured metadata from scientific articles using Machine Learning and Natural Language Processing.This workflow enables the temporal analysis of scientific misconduct and the continuous updating of research databases. The expected outcome of this project is the development of an efficient tool for the preprocessing of scientific articles, contributing to improved information organization and access, as well as the updating of existing databases, such as the one maintained by Recod.AI. (AU)