AI for metadata quality improvement in a CRIS: automating data processes for researchers' benefit

Abstract

The presentation follows up on the update on the CRISalid project delivered at the CRIS2024 conference in Vienna earlier this year and focuses on the challenges posed by the identification of the data on research publications that should feed a CRIS. An eco-friendly, IA-based methodology is proposed to deal with the duplicates that inevitably arise when collecting information on research publications from multiple databases.