Construct Structured Data from Genealogies: Case Study on Kansei Chōshū Shokafu
Satoru Nakamura, Taizo Yamada, Junichi Yamasaki, Yasuo Takatsuki · 2024
This study focuses on the digitization and structuring of the "Kansei Chōshū Shokafu," a compilation of samurai genealogies and family records from late Tokugawa Japan (1799–1812). Specifically, this research introduces a semi-automated structuring process that utilizes a custom layout extraction model along with OCR technology developed by the National Diet Library, Japan. This method, while designed for the "Kansei Chōshū Shokafu," is also applicable to other historical documents with similar structures, thereby enhancing the efficiency of digitization and information structuring processes.