Identifying and extracting authors' Rights Retention Statements from full text academic articles

Many research performing institutes are adopting Rights Retention strategies to help their authors maintain copyright ownership of their work, whilst also enabling broader access and compliance with funder mandates such as Plan S. The implementation of a Rights Retention Strategy offers numerous advantages, including open access assurance, copyright retention, scholarly use regulation, enhanced dissemination, equity promotion, and facilitation of text and data mining. However, the manual incorporation of appropriate rights retention statements into article metadata is labour-intensive and time-consuming. To address this challenge, CORE has co-designed, with repository managers, a machine learning module to automatically identify and extract rights retention statements from full-text articles, streamlining the encoding of this information within article metadata. The integration of CORE services with repository software and the expansion of data extraction capabilities are crucial steps toward promoting a more accessible, transparent and interconnected scholarly ecosystem.