A Quantitative Study on the Linguistic Features of the Tujia Epic Sheba Songs

Previous research on the Tujia epic Sheba Songs has primarily relied on qualitative methods, which are less conducive to objectively examining the texts features. This paper introduces quantitative research methods to explore the textual characteristics of Sheba Songs and compares its features with those of regulated verse, modern Chinese poetry, and Chinese folk songs. The structure of the study is as follows: (1) From the perspective of word frequency distribution, Zipfs law is employed to verify the naturalness of the Sheba Songs text. (2) From the perspective of part-of-speech probability distribution, clustering indices reveal that, compared to regulated verse, the stylistic features of Sheba Songs are closer to those of modern Chinese poetry. (3) Clustering experiments using relevant clustering indices confirm the inclination of Sheba Songs toward the characteristics of folk balladization.