A Quantitative Review of Ancient Chinese Book Digitization: Research Trends and Development Based on DeepSeek-V3 and BER Topic

Li Mingxia; Li Wenke

Vol-10,Issue-6,November - December 2025

Author: Li Mingxia, Li Wenke

Keywords: Ancient books digitization; Deep learning; Large language model; Topic modeling; Digital humanities; Knowledge services

Abstract: This study conducts a comprehensive quantitative analysis of the development and research trends of ancient Chinese book digitization over the past three decades. By employing advanced tools such as CiteSpace for bibliometric analysis, the BERTopic deep learning model for topic modeling, and the large language model DeepSeek-V3 for semantic trend interpretation, the research identifies the developmental stages of the field, eight major thematic areas, and emerging directions such as gamification, knowledge services, and AI model integration. Drawing upon a corpus of 562 peer-reviewed journal articles retrieved from the CNKI database, the study applies both statistical and semantic techniques to uncover key author clusters, topic evolution, and knowledge networks. This work contributes to the understanding of how digital and intelligent technologies are transforming the preservation, dissemination, and reuse of ancient texts, and provides theoretical support for the innovative transmission of traditional Chinese culture in the digital era.

Article Info: Received: 07 Oct 2025; Received in revised form: 08 Nov 2025; Accepted: 14 Nov 2025; Available online: 19 Nov 2025

DOI: 10.22161/ijels.106.17