Page 245 - Journal of Library Science in China, Vol.47, 2021

244
244 Journal of Library Science in China, Vol.13, 2021

as much as possible. Vocabulary reuse needs to select proper classes and attributes from the
existing vocabularies to define the internal and external characteristics of the described objects,
and use shared terms to define the common mode of information exchange. This will realize
the accurate description and formal representation of datasets to improve the interoperability
between datasets, and avoid ambiguity and conflict of expressions. In the process of data
modeling, whether the dataset can be described accurately, whether the connection between data
sets can be established clearly, and whether the human-computer readable and comprehensible
environment can be provided largely depend on the reuse of vocabularies. The paper analyzes
the basic structure of resource description and divides its processing into two phases: data
model and resource annotation. According to the function, it is deeply classified into vocabulary
layer, schema layer, and the data layer. Then it discusses the types and implementation methods
of vocabulary reuse by taking the bibliographic data of the British Library as an example
and shows the differences and applicable environments of various ways, guiding the users of
vocabulary reuse to accurately describe classes, attributes, and value ranges of datasets by using
existing vocabulary, so as to improve the quality of datasets and offer the basis for their further
application. There are two types of vocabulary reuse: vocabulary layer reuse and conceptual
layer reuse occurred at the modeling stage according to different users’ focus. The vocabulary
layer reuse focuses on the selection of vocabularies, and the concept layer reuse focuses on
the matching degree of class and attribute. It is evident that current users prefer the vocabulary
layer to the conceptual layer. With the increasing openness of the vocabularies and sustainable
improvement of its ecological environment, the reuse of the concept layer will be further widely
used. Taking classes and attributes in the data model as the dividing objects, the realization of
vocabulary reuse can be mainly divided into six types: classes or properties reuse, as a superclass/
subclass or subperproperty/subproperty, domain or range, classes operations, inverse properties,
and value restrictions. The choice of vocabulary reuse ways depends on the matching degree of
the target vocabulary and the class and attribute in the data model. Different ways of reuse of the
same vocabulary influence each other and there are issues of class and attribute coordination and
conflict resolution in multiple vocabularies reuse.

Noting the essentials: an explorative tool for catalog annotations in Chinese
rare-book collections
〇a*
LI Hui , CHEN Tao, HOU Junming, LIU Ding, ZHU Qinghua & LIU Wei
Catalog annotations in Chinese rare-book collections, also known as Tiyaos, contain the essential
information regarding a book, e.g., the author introduction, the summary, the nature and style, the

*Correspondence should be addressed to LI Hui, Email:lh9743@126.com, ORCID:0000-0001-7050-1845.