한국고전종합DB 개선방안 II-인물관계망을 이용한 家系圖를 중심으로-

Abstract

This study deals with a method for automatically generating a family tree using 67,000 person data and 200,000 relationship information included in the ‘Interpersonal Relationship Information’ of the Korean Classics DB. This study also shows how to visualize relationships between persons in the created family tree through developed prototype program. Since the printed genealogy was written in vertical Chinese characters, it was common to express the existing family tree in a vertical tree. However, it is shown through logical grounds and experiments that it is effective to use a horizontal tree to digitally express the horizontal writing of Hangul and the family tree. Until now, humanities researchers have spent a lot of time drawing a family tree to understand a person or family line, and there was a lot of waste of time. We crawled the interpersonal relationship data of the Korean Classics DB and build it as an own database, we can automatically create a family tree including one person, and quickly create a family tree in the form the user wants. The addition of a search function helps to quickly intuitively identify relationships. In addition, by linking with character relationship visualization technology developed in previous our researches, it is possible to quickly visually check not only lineal relationships, but also marital relationships and teacher-student relationships. The family tree based on the database depends on the Korean Classics DB data. Although not in this database, it can directly help researchers by creating, correcting, and deleting pedigree information collected by researchers through individual studies. The developed prototype family tree system will become a valuable source technology for future digital genealogy system development.