Loading...

Table of Content

    28 December 2025, Volume 5 Issue 4
    Constructing an Independent Knowledge System for Chinese Digital Humanities from the Perspective of Sovereign AI
    Liu Wei, Liu Shengying, Jin Jiaqin, Shan Rongrong
    2025, 5(4):  3-16. 
    Asbtract ( 37 )   PDF (2031KB) ( 28 )  
    Related Articles | Metrics
    In the context of intensifying global technological competition and the emergence of "Sovereign AI" as a national strategic priority, the development of an autonomous knowledge system has become a central issue in safeguarding national digital sovereignty and cultural security. This paper systematically explores the urgency, theoretical foundations, and practical paths for building China's digital humanities autonomous knowledge system. It first examines the meaning of "Sovereign AI" in terms of technological control and cultural and semantic dimensions, analyzing its role in countering "digital colonialism." Building on this, the paper introduces Professor Duan Yucong's theory of "semantic sovereignty" and the DIKWP (Data-Information-Knowledge-Wisdom-Purpose) model as the core theoretical framework, arguing that the construction of an autonomous knowledge system in China must involve deep leadership over both the content and value intentions of knowledge. By analyzing key cases such as the China Academic World platform, the Shidian Ancient Books platform by ByteDance and Peking University, and the ICH-Qwen large language model, the paper reveals a construction path that integrates top-level design, technological empowerment, and paradigm innovation. The research indicates that China's digital humanities autonomous knowledge system is built on the Sovereign AI strategy, with semantic sovereignty as the core, using digital platforms and intelligent tools to achieve data autonomy, knowledge innovation, and value guidance. Finally, the paper reflects on potential challenges such as the "sovereignty trap" and envisions the future of an open, confident, and globally influential Chinese digital humanities academic community.
    Bringing Local Voices to the World: A Seminar on the Glocal Practice of Sound Archives—A Review on Their Creative Reconstruction in Digital Humanities
    Wu Yang, Liu Ying, Wan Zhenyan
    2025, 5(4):  21-25. 
    Asbtract ( 22 )   PDF (1191KB) ( 22 )  
    Related Articles | Metrics
    Sound archives, serving as vital carriers of cultural heritage,present enhanced possibilities—including data storage and migration, cataloguing, annotation, implementation, and dissemination—through the application of digital humanities technologies. This interdisciplinary symposium, “The ‘Glocalisation’ Practices of Sound Archives,” explores the integration of sound archives with digital methods for their preservation,interpretation,and activation. Contributions include Professor Richard Wolf’s reflections on ethnomusicological film as a medium for archiving and contextualizing traditional music; Dr. Xiaoshi Wei’s emphasis on deep description and recontextualisation of private sound collections; Dr. Matthew James’s presentation on the AI-enhanced, globally-oriented “Echo Arc” project for audiovisual archives; and Curator Colin Chinnery(Qin Siyuan)’s discussion on the cultural and technical dimensions of sound collection and exhibition. Together, these case studies illustrate current interdisciplinary innovations and pathways in sound archive research, while underscoring the need for future digital practices to preserve cultural diversity better and ensure that sound is accurately heard, understood, and activated within a globalised framework.
    The Temporal Dimension of Ethnographic Film: Cases from the Wakhan Corridor and South India
    Richard K.Wolf
    2025, 5(4):  26-34. 
    Asbtract ( 22 )   PDF (5099KB) ( 20 )  
    Related Articles | Metrics
    Based on Richard K. Wolf’s presentation at the symposium “The ‘Glocalisation’ Practices of Sound Archives,” this article delves into the reconstructive practices of archival materials across different temporal dimensions in ethnomusicological film. By comparing the disparities in traditional music archive development across different countries and regions and their resulting realities, it emphasises the core value of field recording archives and systematic metadata documentation. Using two of his own ethnomusicological films as examples, Professor Wolf elaborates on their role in activating and reconstructing sound archive materials. This approach not only enhances the preservation of sound and imagery but also, through re-contextualization, helps communities reflect on and engage with their own history while pondering the relationship between digital technology and cultural heritage.
    Presenting Folk Sound Documents in Museums
    (Colin Siyuan Chinnery
    2025, 5(4):  35-47. 
    Asbtract ( 24 )   PDF (17086KB) ( 19 )  
    Related Articles | Metrics
    Based on Colin Siyuan Chinnery's report presented at the symposium “The ‘Glocalization’ Practices of Sound Archives,” this article delves into the practical pathways for inheriting and revitalizing folk sound archives with in the museum context. Through the archaeological investigation, observation, salvage, and documentation of sound categories such as “long-vanished sounds,” “ soon-to-disappear sounds,” and “ socially enduring sounds,” it explores how such sound archives can transcend mere documentary preservation via museum narrative techniques. This approach serves as a unique methodology for understanding social transformation and diagnosing cultural ecology, thereby linking individual memory with collective history.
    Beyond Digitisation: Recontextualising Sound Archives in China
    Wei Xiaoshi
    2025, 5(4):  48-59. 
    Asbtract ( 22 )   PDF (8442KB) ( 18 )  
    Related Articles | Metrics
    This article examines the paradigm shift in the preservation and study of sound archives in China, particularly those of traditional musics, moving from a focus on technical digitisation towards annotation, documentation, and—in its key concept—recontextualisation. Through multiple case studies—from early wax-cylinder recordings to contemporary field collections—I demonstrate how sound archives can be recontextualised by excavating their narrative, emotional, and cultural dimensions. I argue that sound archives should be treated as living cultural texts, embedded with rich layers of social,historical,and personal meaning. Such an approach reactivates sound as a medium for cultural dialogue, identity negotiation,and interdisciplinary research. Ultimately, the study calls for a more integrated, multimodal methodology in digital humanities—one that engages with sound archives not as isolated artefacts,but as dynamic resources for understanding “where we are” and “who we are.”
    To Augment, Not Replace: AI Metadata Generation Models for Audiovisual Archives and Cultural Interpretation
    Wei Xiaoshi, Matthew James
    2025, 5(4):  60-67. 
    Asbtract ( 16 )   PDF (1310KB) ( 21 )  
    Related Articles | Metrics
    This article addresses the challenges faced by GLAM institutions in managing sound and audiovisual archives, characterised by exponential digital growth, funding constraints, and a shortage of specialised cataloguing expertise. In response to this “triple dilemma,” we advocate for developing AI-assisted metadata tools designed to augment—not replace—human expertise, thereby shifting focus from digitisation to knowledge organisation and recontextualisation. Through a case study on early 20th-century Chinese quyi(narrative singing) recordings, we demonstrate how a domain-specific AI model—integrated with a Retrieval-Augmented Generation(RAG) architecture and trained on classical texts and expert annotations—enables deeper semantic analysis and culturally sensitive description. Ultimately, we call for collaborative development of such AI systems among ethnomusicologists, archivists, and technologists. This human-in-the-loop approach aims to enhance the global accessibility and interpretability of sound archives while preserving the accuracy and richness of cultural contextualisation.
    Construction and Application of the Knowledge Base for Duan Yucai's Shuowen Jiezi Zhu
    Shen Xiaoni, Peng Weiming, Hu Jiajia
    2025, 5(4):  68-83. 
    Asbtract ( 23 )   PDF (2662KB) ( 25 )  
    Related Articles | Metrics
    Duan Yucai's Shuowen Jiezi Zhu(Annotations on Shuowen Jiezi) stands as the pinnacle of research on Shuowen Jiezi, centrally reflecting the academic achievements of the Qian-Jia School. Currently, digitalization efforts for this work have largely remained at the textualization stage, with limited progress in the in-depth exploration and systematic presentation of its knowledge system. This research constructs a knowledge base using this work and its draft manuscript Shuowen Jiezi Du(Reading Notes on Shuowen Jiezi) as foundational texts. It designs a three-tier classification system encompassing five major knowledge categories, ten knowledge sets, and fifty-six knowledge points, completing the annotation and structured representation of tens of thousands of knowledge instances. An interactive platform featuring original text retrieval, knowledge navigation, and annotation management functions has been developed. This study represents a systematic practice of digital humanities methods in the field of traditional philology,providing crucial references for the in-depth digitization of ancient texts and the innovation of humanistic research paradigms.
    An Ambiguous Reference Resolution Method for Party History Literature Integrating Semantic Understanding and Knowledge Graph Reasoning
    Ran Lingyu
    2025, 5(4):  84-98. 
    Asbtract ( 21 )   PDF (1903KB) ( 20 )  
    Related Articles | Metrics
    The intelligent processing of Party history literature faces significant challenges due to the extensive use of pseudonyms,  alternative designations, and complex implicit relationships. This study proposes a multi-strategy semantic understanding and dynamic knowledge graph reasoning-based method for ambiguous reference resolution to address three major challenges in this field: the semantic gap, temporal evolution, and sparse evidence. The method constructs a domain-specific lexicon covering over ten thousand entities and a pseudonym-real name mapping database to incorporate prior knowledge. A domain dictionary-guided negative sample sampling strategy is employed to fine-tune pre-trained language models, enhancing their semantic perception of specific expressions. Finally, a time-constrained graph neural network reasoning algorithm is applied on a self-built temporal knowledge graph to mine implicit relationships and perform consistency verification. Experimental results demonstrate that the proposed method achieves an overall F1 score of 80.6% on authoritative evaluation metrics, significantly outperforming existing baseline models,and effectively uncovers deep historical correlations. The research outcomes have been integrated into a visual prototype system, providing a reliable intelligent tool for Party history research.
    Exploring the Possibilities of Literary Reception Studies through Quantitative Analysis: A Case Study of Matsuo Bashō’s Reception in the Modern Era
    Hibi Yoshitaka, Jiang Hui
    2025, 5(4):  99-117. 
    Asbtract ( 17 )   PDF (2906KB) ( 25 )  
    Related Articles | Metrics
    This study aims to explore the potential of digital humanities methods in literary history research through a quantitative analysis of the reception of Matsuo Bashō in modern haiku. Utilizing the digital collections of the National Diet Library of Japan, the frequency of citations of Bashō’s haiku in literature from the Meiji to the prewar Shōwa period was statistically analyzed. Additionally, the Jaccard coefficient was employed to conduct a similarity analysis of large-scale haiku collections from the Meiji, Taishō,and prewar Shōwa periods. The research found that Bashō’s influence on haiku gradually declined over time, while the haiku of Masaoka Shiki, known for his criticism of Bashō, exhibited a style closer to Bashō’ s than that of general haiku from the Meiji period. This demonstrates that computer-based quantitative analysis can capture subtle expressive differences that are difficult for humans to detect,extracting the “unconscious rhetoric” of individuals and eras. Digital humanities methods can reveal deep-seated patterns that traditional research struggles to access,providing new pathways for the construction of “digital literary history”.
    Artificial Intelligence's Involvement in the Path of Film Criticism: “Non-human” Interpretation, Algorithmic Illusion and Platform Orientation
    Song Wei
    2025, 5(4):  118-128. 
    Asbtract ( 27 )   PDF (1737KB) ( 25 )  
    Related Articles | Metrics
    In the era of cultural production increasingly shaped by artificial intelligence, big data and algorithmic logics have become deeply embedded in the fabric of film-criticism practices. Within this process,“AI distant viewing” conducted under the framework of computational film studies, together with algorithm-driven computational understanding and generative criticism,does not in fact construct a complete chain of “viewing-interpretation-critique. ” Instead, it tends to produce a de-aestheticized and de-experiential “meaning-vacuum” form of approximation. At the same time, platform-oriented aggregation and recommendation based on probabilistic statistics merely fabricate a service-oriented feedback mechanism that resembles criticism and appears to generate critical efficacy,while its essence lies in a field of homogenized opinion reproduction governed by algorithmic preset rules. The transformation of film criticism in the age of AI is therefore not a paradigmatic overhaul but a path-dependent intervention that introduces heterogeneous forces into processes of viewing,interpretation, and dissemination,  thereby exerting multidimensional infiltration, redirection, and structural constriction upon existing critical paradigms.