EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Construction of Thangka Cultural Knowledge Graph and Cross-regional Inheritance Application

Hao Zhang, Ramin Hajianfard

Journal of Humanities and Cultural Studies · 27.05.2026 · 1 (2) · 37-41

DOI: 10.54097/wztmad74 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Humanities and Cultural Studies, online erschienen am 27.05.2026. Verfasst von Hao Zhang und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Thangka - sometimes transliterated thang-ka or tangka - is a centuries-old Tibetan Buddhist painting tradition on cotton or silk, encoding layered iconographic meanings accessible only through initiated ritual and artistic training. UNESCO inscribed Thangka painting on its Representative List of the Intangible Cultural Heritage of Humanity in 2009. That inscription, while conferring international recognition, also exposes a vulnerability: trained masters are declining in number, and younger audiences increasingly encounter these paintings through digital interfaces rather than in shrine settings that give them their full cultural significance. How to systematically represent, preserve, and transmit Thangka across regions has become pressing. Over the past decade, digital preservation of intangible cultural heritage (ICH) has drawn sustained attention. High-resolution imaging, 3D scanning, and online museum archives have produced vast digital corpora. Yet most efforts treat digitized Thangka as static image resources, disconnected from the rich semantic networks - iconographic attributes, artisan lineages, historical trajectories, ritual functions - that make Thangka culturally intelligible. Without structured semantic representation, digital collections remain fragmented and resist being queried, interconnected, or reused for education and inheritance work. Knowledge Graph (KG) technology, grounded in semantic web and linked data principles, offers a way forward. By representing cultural heritage entities and their relationships as nodes and edges in a graph structure, KGs enable semantic reasoning, cross-dataset linking, and intelligent question-answering. This paper presents the construction of a domain-specific knowledge graph for Thangka cultural heritage and evaluates its application to cross-regional inheritance. The KG integrates multi-source heterogeneous data - museum collections, textual sources, artisan interviews, and online archives - and formalizes the domain ontology in OWL, aligned with CIDOC CRM. A prototype platform, the Thangka Cultural Knowledge Graph Platform (TCKGP), demonstrates the KG's utility across three scenarios: museum digital curation, online artisan training, and transnational cultural education. Evaluation results show that the KG approach substantially outperforms conventional relational database approaches in semantic retrieval tasks and effectively supports cross-regional inheritance practices. This study contributes to digital heritage scholarship by providing a reusable ontology and a practical methodology (the TCKG Model) for applying knowledge graphs to ICH, and to Tibetan Buddhist art studies by supplying the first formally specified, machine-readable ontology for Thangka iconography and lineage.

Zitieren

APA

Zhang, H., & Hajianfard, R. (2026). Construction of Thangka Cultural Knowledge Graph and Cross-regional Inheritance Application. Journal of Humanities and Cultural Studies, 1(2), 37-41. https://doi.org/10.54097/wztmad74

Vancouver

Zhang H, Hajianfard R. Construction of Thangka Cultural Knowledge Graph and Cross-regional Inheritance Application. Journal of Humanities and Cultural Studies. 2026;1(2):37-41. doi:10.54097/wztmad74

BibTeX
@article{Zhang2026,
  author = {Zhang, Hao and Hajianfard, Ramin},
  title = {Construction of Thangka Cultural Knowledge Graph and Cross-regional Inheritance Application},
  journal = {Journal of Humanities and Cultural Studies},
  year = {2026},
  volume = {1},
  number = {2},
  pages = {37--41},
  doi = {10.54097/wztmad74}
}
Kostenlos starten →