Publications
Detailed Information
군집분석 기법을 이용한 텍스트의 계통 분석 : 수궁가 고고천변 대목을 대상으로 : A Study of Measuring Text Distances using the Hierarchical Clustering Method in Application to Pansori Narratives
Cited 0 time in
Web of Science
Cited 0 time in Scopus
- Authors
- Issue Date
- 2009
- Publisher
- 서울대학교 인문학연구원
- Citation
- 인문논총, Vol.62, pp. 203-229
- Keywords
- 텍스트 거리 ; 레벤시타인 거리 ; 최소편집거리 ; 군집 ; 분류 ; 계통분기 ; 텍스트 마이닝 ; text distance ; Leveshtein distance ; minimum edit distance ; cluster ; classification ; phylogeny ; text-mining
- Abstract
- In this study, a computational method was used to measure the distance
values between texts, especially to measure the variations which took place
during the transmission of texts. Gogochenbyeon (a scene of rabbit
marvelling at the beautiful scenary of land world) is called nundaemog
of Sugungga. Nundaemog is one of the most essential and important
parts of a Pansori, like arias of an opera. So many Pansori singers recorded
those nundaemog, and therefore, gogochenbyeon has been recorded by
several famous singers. We selected 16 albums of 16 singers, and
transcribed the narratives from those albums to the texts, and also put the
musical expressions (rhythm, rhythm pattern, key, beat) on the Korean
traditional musical score Jungganbo. From these raw materials, the
text-distances of the transcribed narratives were measured using Levenshtein
distance, and the distances were normalized by the length of the narrative
strings. And then the texts were clustered using hierarchical clustering
methods and turned into the phylogenetic trees using neighbor-joining
methods. After the result, we found some differences between the lineage of the Pansori transmission and the real narratives that some singers like
Park Tongjin and Song Sunseob are somewhat distant from their pupils or
masters.
- ISSN
- 1598-3021
- Language
- Korean
- Files in This Item:
Item View & Download Count
Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.