Topic Mining of Modern Poetry For Digital Humanities: Case Study of Gu Cheng Poetry

Le résumé fourni par la source

Using topic extraction and visualization method to explore the themes of Gu Cheng's complete poems, which provides a reference for the quantitative interpretation and analysis of modern poetry works. According to Gu's self-staging, we divide Gu's complete poems, and use Latent Dirichlet Allocation (LDA) model and word cloud visualization technology to conduct staged topic extraction and visual analysis of Gu's poems. Finally, we divide Gu's poems creation into four stages of nature stage, culture stage, counter-culture stage, and non-self stage. On the staged topic extraction experiment of Gu's poems, we obtain four stages of topic extraction results and word frequency cloud maps, and in-depth analyze the theme changes and reasons of Gu's poems at different stages. Results show that the used topic extraction and visualization method could objectively and comprehensively reveal the content and change process of Gu's poems.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.