[1N4-OS-18] OS-18

[1N4-OS-18-02] Introduction to research on Chinese text mining in the field of history

北魏墓誌の銘辞に対するテキストマイニング分析

〇Seiko OCHI¹ (1. Meijo university)

Keywords:Textmining, classical chinese

Traditional historical research does not use quantitative analysis methods, but instead cites a few historical sources and generalizes based on their characteristics. This method involves determining whether the cited historical material happened to catch the researcher's attention, or whether it was selected as typical after examining a large amount of historical material, or whether it was selected as typical. However, the problem is that it remains unclear why this can be said to be typical. Therefore, it is necessary to verify the results in a comparative and falsifiable manner through quantitative analysis, which is a necessary task since history is a humanities science. This report uses text mining, which is specialized for text analysis among data mining, to present research results in the field of history. Specifically, I used the analysis tools KH Coder and ChineseTextProject to clarify human networks and gender stereotypes that were not visible through conventional research, using the analysis tools KH Coder and ChineseTextProject for Chinese texts (inscriptions on epitaphs from the Northern Wei Dynasty in China).