Page 165 - Journal of Library Science in China, Vol.47, 2021

164 Journal of Library Science in China, Vol.13, 2021

presented by normative referents such as associated datasets, thesauruses, thesaurus lists or
[28]
taxonomies from different fields . Finally, the entity semantic tags or entity portraits in different
dimensions are formed, and the multi-dimensional nuclear aggregation effect generated around
such portrait tags is the “dynamic aggregation” process in the archival data organization, which is
of great significance to the archival data characterization of attributes such as credentials, integrity
and credibility.

2.3.3 Storytelling: narrative representation of archival data
Compared with “discovering” and “reorganizing”, the “storytelling” of archival data belongs to
the scope of knowledge services, that is, the archival data that has been organized in multiple
dimensions is sorted out into a book and mapped to a multi-dimensional visual space, and the
memory contained in the archival data is narratively represented in the form of a theme story. This
process includes “user probing”, “structure definition” and “story statement”.
(1) User probing. Traditional data or knowledge storytelling can be divided into two modes:
[29]
creator-driven and audience-driven . For archival data, especially archival data that is currently
declassified and ready for development, the identities of creator and audience are developing
a convergence relationship in a progressive way. In the postmodern concept of archival
science, archives themselves are produced by the public, representing the entire process from
individual memory to group memory, and finally rising and condensing into an organic part
of national memory. To a certain extent, archive users are not only producers of archival data
but also the managers, organizers, communicators and consumers, who play a key role in the
storytelling process of archival data. At present, the archival circles at home and abroad are
slightly macroscopic in the research perspective of archival users, and the probing and mining
of user intention, behavior, psychology, experience, emotion and other elements are insufficient.
Therefore, the storytelling of archival data not only needs to pay attention to “signifier” directly
presented in the data, but also needs to dig deep and sort out the “signified” contained therein. At
the same time, archival data storytelling needs to further define the template rules, story themes,
telling forms, organizational plans and other contents of archival data storytelling.
(2) Structure definition. The story structure represents the basic framework and mode of archival
data storytelling. After realizing the shallow disclosure and deep calculation of the memory entity,
it is necessary to consider the subjective factors such as the user’s knowledge requirements and
the narrative scheme designed by the researcher so as to provide humanistic explanations for the
associated archival data. In terms of the structure of data storytelling, CHAO Lemen summarized
the five-step narrative structure and maturity curve model proposed by Aristotle and divided the
[30]
data story structure into five stages: introduction, rise, climax, descent and ending . This type of
structure is relatively macroscopic as a whole, and the organization and arrangement of embedded
elements have not yet been discussed. This paper summarizes the story structure that can be used
for archival data organization from the perspectives of domain and structure (see Figure 4).