Metadata is as important as materials in building a digital archive of valuable materials and books. Metadata includes information such as the author, year of publication, and history of the material or book, and provides the basis for its academic value.
However, in order to realize the above in the current digital archiving system, the following issues must be addressed in order to search and use academic resources with backgrounds in various fields of expertise on the same platform: (1) Currently, the judgment of experts is necessary, and therefore, similar to the issues faced by annotation in machine learning, it is difficult to provide metadata for the digitization of materials. (2) Libraries, archives, and museums have different metadata assignment policies based on their respective backbones (library and information science, archives science, and museology) and are unable to establish a unified metadata format that transcends the boundaries of their respective fields. Therefore, it is not possible to establish a unified metadata format that transcends the boundaries of each field.
In order to solve the above problem, this study will explore the possibility of metadata assignment using generative AI under the framework of an international joint research project. We will then construct an international model for the knowledge base of digital academic space and show how the Japanese humanities should proceed in the DX-AI era.
