Abstract:

Tang dynasty poetry is a cultural treasure of China， with a large number of works， diverse styles and themes. In order to discuss the characteristics of Tang dynasty poetry in different periods， this paper used natural language processing （NLP） and statistical methods such as K-means++ clustering analysis， repeated measures ANOVA， and paired sample t-test to analyze the stylistic characteristics and differences of poetry in the Early Tang， Golden Age of Tang， Middle Tang， and Late Tang periods.The results showed that there were significant differences in the poetic styles between the Early Tang and Golden Age of Tang periods， while the differences between the Middle Tang and Late Tang periods were relatively small.

Key words:

NLP, K-means++ clustering analysis, ANOVA, Tang dynasty poetry, feature analysis