CNMARC数据中作品的识别与建模  

Identifying and Modeling Works in CNMARC Bibliographic Records

作　　者：许磊[1] Xu Lei(Shanghai Library)

机构地区：[1]上海图书馆

出　　处：《图书馆杂志》2022年第1期67-74,121,共9页Library Journal

基　　金：国家社会科学基金项目“面向数字人文研究的图书馆开放数据体系构建与服务模式设计研究”(项目编号:18BTQ027)的研究成果之一。

摘　　要：作品识别是书目数据重构的关键的一步,并决定了所有的后续操作。书目世界的作品概念继承自潘尼兹,经过柳别斯基等人的阐述,最终确定了书目数据作品与载体的二分法。可操作性的作品识别可以在具现化的文本层面实现。指称作品的题名、所有署名的责任者,以及基于用户研究确认的语种共同构成识别作品的基本元素。同时,在作品模型中引入顶层的抽象实体——超级作品,实现对相关作品的聚合,并最终形成作品家族。随后,根据语种、汇编类型对CNMARC数据进行分包,由简入难,依次对每类数据制定相应的作品识别方案。并以非翻译类单语种文献、翻译作品和丛编作品为例对CNMARC数据中作品的识别与建模进行说明。Work identification is a key step in bibliographic data reconstruction and determines all subsequent operations.The concept of the work in the bibliographic universe was inherited from Panizzi,and the dichotomy of the work and the manifestation of the bibliographic data was finally established after the elaboration of Lubetzky and others.Operability works can be identified at the embodied text level by referring to the title of the work,all those contributors,and the language identified based on user research.At the same time,the top-level abstract entity—Superwork is introduced into the work model to realize the aggregation of related works and finally form the work family.CNMARC bibliographic records needs to be subcontracted according to the language and the type of aggregates.From simple to difficult,the corresponding work identification programs are developed for each type of records.The identification and modeling of works in the CNMARC records are illustrated by non-translation monolingual literatures,translated literatures and series.

关 键 词：作品 CNMARC BIBFRAME 关联数据 

分 类 号：G254.3[文化科学—图书馆学]