Page 174 - Journal of Library Science in China, Vol.47, 2021

NIU Li, GAO Chenxiang, ZHANG Yufeng, YAN Shi, XU Yongjun & LI Anrunze / Discovering, reorganizing 173
and storytelling: paths and methods of archives research on the perspective of digital humanities

by International Council on Archives Expert Group on Archival Description (ICAEGAD) [38] ,
combined with the previous metadata annotation and entity recognition results, the definition and
annotation of the context relationship of photo archives are completed. Among them, at the internal
context level of the photo archives, based on the key entities mentioned by RiC-O such as object,
person time, place, event, etc., we have completed most of the work at the metadata and entity
recognition level, and then combined with the analysis of WU Baokang’s Academic Chronicle, the
field research of the archives, and the access to archives to fill in the specific examples missing in
each photo, and examine their relationships and corresponding relationship perspectives. In terms
of the external context, that is, the relationship between individual photographs, under the premise
of respecting the existing logic of archival arrangement [39] , we cut from different dimensions
such as time and theme, clustered several photos in the sequence in context, and built a specific
narrative framework to form the infrastructure of archival data storytelling, the process is shown in
Figure 8.

Figure 8. Hierarchical context recognition of WU Baokang’s photo archive

4.2.2 Cascading “reorganizing” of archival data
In the “reorganizing” section of archival data, we design an ontology model including “entity
set”, “type” and “constraint” three-level conceptual set with reference to the SEM model.
Among them, at the level of “entity set”, the concept of “sem: actor” is concretized into
two concepts of “FOAF: Person” and “FOAF: Organization”, and the concepts of “CRM:
Document” and “CRM: Physical Thing” are newly added to reflect the thesis work or physical
thing mentioned in photos or texts, and the semantic association relationship between the new
concepts is supplemented on the basis of the established object attributes of the SEM model, and