From Weaving Patterns to Data Patterns: AI-Driven Cultural Documentation of East Sumba’s Ikat Traditions

Laely Indah Lestari, Evi Novianti, Dadang Sugiana and Ute Lies Siti Khadijah

Data and Metadata, 2025, vol. 4, 1129

Abstract: Introduction: This study aims to develop and test a metadata-driven, AI-assisted framework for documenting the visual, oral, and symbolic elements of ikat weaving in East Sumba. It seeks to explore how artificial intelligence can transform traditional knowledge into machine-readable cultural data structures while maintaining epistemological integrity and community participation. Methods: This research employed a hybrid qualitative-technical methodology. Ethnographic fieldwork was conducted with ikat artisans in East Sumba to gather narrative and visual data, including interviews, ritual transcripts, and photographs of woven fabrics. These data were analyzed using a combination of natural language processing (NLP) and computer vision algorithms. NLP was used to extract recurring linguistic patterns and cosmological themes, while computer vision categorized visual motifs by type, symmetry, and symbolic meaning. Results:The AI-driven approach effectively captured the symbolic and narrative complexity of the ikat tradition. Computer vision techniques successfully identified and classified motif types and regional styles, linking them to spiritual meanings conveyed by the artisans. NLP analysis of transcribed interviews revealed consistent narrative patterns related to ancestral cosmology, customary law, and motif symbolism. Conclusion: This research demonstrates the viability of combining AI technologies with ethnographic fieldwork to create a robust, ethical, and culturally sensitive system for documenting living traditions. By translating the complexity of ikat knowledge into semantic data patterns, the study provides a model for intelligent cultural heritage documentation. The key contribution lies in bridging indigenous epistemologies with digital infrastructures, enabling scalable cultural preservation without compromising authenticity.

Date: 2025
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:dbk:datame:v:4:y:2025:i::p:1129:id:1056294dm20251129

DOI: 10.56294/dm20251129

Access Statistics for this article

More articles in Data and Metadata from AG Editor
Bibliographic data for series maintained by Javier Gonzalez-Argote ().