조선시대 科試 자료와 시맨틱 데이터 모델링

Abstract

The Study on Building the DB System of Civil Service Examinations in the Joseon Dynasty is a research project designed to create a comprehensive database cataloging materials related to the civil service examination of the Joseon Dynasty. This database aims to establish a foundation for understanding the development of academic, literary, and publishing cultures in the Joseon Dynasty, while also analyzing the history of a rhetorical style and a way of constructing logic within the Civil service examination system. This paper addresses the key scholarly concerns raised by the incorporation of data-processing technologies into research, elaborating on the academic tasks that follow. It focuses on proposing the academic utility of data-driven approaches to the study of civil service examination materials. The civil service examination database has the task of connecting various information units centered around the event of the ‘examination’, linking elements such as questions, answers, writing styles, certificates, rosters, official historical records, collected works, reference books and manuals, statute books, unofficial tales, diaries, letters, examination systems and terms, person, location, and time. To effectively organize the semantic relationships across diverse materials related to the civil service examination, this study employs a semantic data model-based design. It aims to implement a knowledge graph by defining various humanities-related entities with different contextual origins and characteristics as nodes and connecting them through semantic definitions as edges. When building datasets for a single type of textual material, it is common to design data in a way that meticulously conceptualizes the structural features of its form, content, and genre. In contrast, the civil service examination dataset does not focus on narrowly conceptualizing specific material groups but rather emphasizes connecting fragmented materials related to civil service examinations, particularly those centered around questions and answers, to restore historical contexts and supplement bibliographic limitations. By examining and linking examination infor-mation from various sources, including rosters, examination papers, collected works, and collections of examination articles, both directly and indirectly related to civil service examinations, this study seeks to address gaps in existing knowledge and proposes a data model that reconstructs and expands the temporal and spatial contexts of civil service examinations. This study prioritizes the design of categories related to civil service examinations, focusing on core elements such as Exam, Time, Location, Question, Answer, Sigwon(試券), Person, Work, Collection, Bangmok(榜目), Collection of examination articles, and Certificate. These elements are established as fundamental Classes, and their Attributes and Relations are systematically designed to illustrate how diverse information related to the examinations can be conceptualized and interconnected within a knowledge graph framework. During the data modeling process, Triple(S-P-O) data were utilized to implement a data network in the form of a knowledge graph, visually confirming the validity of the data model proposed in this study.