The rise of digital humanities has highlighted the need for more intuitive methods to explore ancient literary works. Traditional approaches often fail to fully capture the depth of these texts, particularly their cultural and structural layers. This paper proposes a visualization framework that uses multi-scale feature extraction and graphic-semantic mapping to transform complex literary elements into interactive visual representations. The framework involves a model that extracts key features—such as thematic intensity, imagery density, and emotional trajectory—and a semantic mapping algorithm that converts textual networks into graphical structures. Interactive visualization, with customizable views and dynamic filtering, enhances both reading experience and scholarly analysis. Experimental results show improvements in information retrieval speed and comprehension. The paper concludes by discussing potential applications and future improvements, contributing valuable insights to the field of digital humanities.
D.Z. Shen (Wed,) studied this question.