Two approaches to using AI in cataloguing

a webinar sponsored by the IFLA Artificial Intelligence Special Interest Group (AI SIG). Cataloguing is traditionally a manual act of metadata creation that takes substantial focused person-hours. Conventionally, libraries have relied on several approaches to increase efficiency of metadata creation, such as sharing of records via z39.50 protocol or via cooperatives like OCLC. Original cataloguing, however, remains a very time-consuming task, the duration of which depends on a title’s complexity. SGCAT Hui Ling, National Library Board of Singapore To that end, the National Library Board of Singapore developed a custom GPT prototype (“SGCAT”) to streamline bibliographic metadata creation and enhance efficiency in delivering library materials to patrons. Powered by OpenAI GPT, SGCAT is customised to follow specific cataloguing rules and local library practices, effectively serving as a smart cataloguing assistant for our cataloguers. To ground it in fact, SGCAT pulls relevant bibliographic data from trusted sources such as NLB’s vendor-provided order information and Google Books. Through rigorous prompt engineering, SGCAT is currently able to draft a MARC record from a single ISBN in seconds, potentially speeding up cataloguing by at least 2x in combination with human review. These assistive capabilities help transform the cataloguer’s role from creator to reviewer, freeing them to focus more on cerebral tasks like subject cataloguing and record refinement. The team is currently improving this tool and working on incorporating new order information from vendors and selectors via our new acquisition systems to further refine the process. Annif Mona Lehtinen, National Library of Finland This webinar also introduces the automatic subject indexing tool Annif with use case examples. Speakers from the National Library of Finland will discuss the basics but are ready to discuss more advanced questions and other aspects of automated metadata generation. – such as automated extraction of bibliographic metadata and automated description of images – with Annif developers.