Knowledge Organization Systems

The Knowledge Organization Systems (KOS) Research Unit focuses on document management throughout their lifecycle, terminology management, knowledge organization, and semantic interoperability in various sectors and contexts, such as digital healthcare, cybersecurity, and cultural heritage. The KOS Research Unit's research team is committed to supporting the Ministry of Health and the Agency for Digital Italy (AgID) in the development and implementation of the Electronic Health Record (EHR), focusing in particular on semantic aspects related to the management of clinical data and documents.

The KOS Research Unit also hosts the LOINC® Italia Working Group. IIT is, in fact, the Italian partner of the Regenstrief Institute and, through the LOINC® Italia Working Group (Organizational Measure of the IIT Director no. 9/2013 of 31/10/2013), is responsible for the development and updating of the Italian version of LOINC® and the translation of related information materials into Italian. To support industry professionals in the implementation and use of the LOINC coding, the LOINC® Italia Working Group: organizes training seminars on the standard and on the use of the RELMA search and mapping software; offers a Mapping Validity Check service; provides ongoing assistance, both for informational and more specifically technical requests, through the online helpdesk. The main activities of the LOINC® Working Group are: representing CNR as the official Italian partner of Regenstrief Institute Inc.; maintaining the Italian version of LOINC® and all related materials; supporting Italian users of LOINC®; managing and maintaining the online helpdesk www.loincitalia.it; Improving interoperability between Italian local systems. In the context of the protection and enhancement of cultural heritage, the Unit integrates Artificial Intelligence methodologies and tools to increase data quality, scalability, and reusability.

Specifically, it applies AI models (including Large Language Models) and Handwritten Text Recognition (HTR) techniques to the semi-automatic transcription of manuscript texts, the extraction of entities and concepts, and the identification of semantic relationships between concepts. The goal is to produce structured, disambiguated, and interoperable data, reduce ambiguity, and improve interpretability and quality control. Regarding the long-term preservation of historical documents, the KOS Research Unit focuses on their recovery, cleaning, anoxia disinfestation using a nitrogen tent, digitization, and vacuum preservation. The process has been successfully applied to archival material from the former psychiatric hospital in Reggio Calabria, as part of the "Carte da legare" project funded by the Ministry of Cultural Heritage and Activities and Tourism. It has also been extended to the digitization, description, and valorization of the historical archives of Girifalco. In both contexts, digitization is accompanied by systematic work on metadata and knowledge organization systems (controlled vocabularies and ontologies) and the use of AI/HTR to transform documentary collections from simple consultation tools into true research infrastructures: searchable corpora, linkable to entities and concepts, and reusable in historical, social, and interdisciplinary studies.