Page 176 - Journal of Library Science in China, Vol.47, 2021

NIU Li, GAO Chenxiang, ZHANG Yufeng, YAN Shi, XU Yongjun & LI Anrunze / Discovering, reorganizing 175
and storytelling: paths and methods of archives research on the perspective of digital humanities

Figure 10. “Dynamic aggregation” of WU Baokang’s photo archive data knowledge graph

4.2.3 Fusion “storytelling” of archival data
At the level of “storytelling” realization of WU Baokang’s archive data, the team develop a
knowledge service system based on B/S architecture as a user interaction platform, which sets up
user opinion survey and user message modules, and gives users the permission to complete the
narrative of WU Baokang’s photo archives according to the presentation content and methods of
different theme photos.
In the early research on the storytelling of WU Baokang’s photo archives, we have not yet
applied a fully automated narrative mechanism based on deep learning model, but adhered to the
humanistic research concept of later generations to reconstruct the memories of predecessors based
on first-hand historical materials, and adopted a crowdsourcing narrative method to guide users to
realize multi-perspective narration of archival stories based on object detection results, contextual
information and knowledge graphs of archival data [40] . In Figure 11, we build a semi-automated
archival data storytelling framework, select five photo archives from the period when WU Baokang
was transferred to Beijing in the early 1950s to create the Archives Department of Renmin
University of China, and process the objects one by one, and then use the previously constructed
and stored character semantic knowledge graph as the key context information to create a basic
narrative space for users and provide necessary elements, and also complete the limited definition
of the narrative theme and scope. In order to accommodate the user’s subjective thinking on the
premise of following the basic objective facts, the same historical facts are described in multiple
dimensions.