Preserving the Pulse of Bali: A Semantic Approach to Digital Heritage

TL;DR

To prevent the erosion of Balinese cultural knowledge, researchers have built a cloud-based digital portal that uses a specialized kulkul ontology and community crowdsourcing. Unlike simple digital archives, this system allows users to perform complex semantic searches (e.g., "Find kulkuls used for specific rituals with exact physical dimensions"), ensuring that the intangible traditions of the kulkul (traditional wooden bells) are preserved as living data.

Background: Beyond the Physical Object

Cultural heritage is more than just ancient artifacts in a museum; it is a continuous tradition lived through daily practices. In Bali, the kulkul system serves as a traditional communication tool, yet its nuances vary across villages and are at risk of being forgotten by younger, tech-savvy generations. The primary challenge in digitizing such heritage is capturing context—the "why" and "how" behind the object.

Methodology: The Power of Semantic Structure

The researchers didn't just build a website; they built a Knowledge Model. By using the Tri Hita Karana (a philosophy of harmony) to classify data, they moved beyond simple keyword searches.

The Semantic Search Architecture

The portal utilizes an Ontology-based approach, allowing for relationships like thk:isPartOf or thk:isUsedFor. This enables the system to handle complex logic through SPARQL queries.

Figure 1: The evaluation and recruitment process highlighting the integration of user feedback into the semantic portal.

From Crowdsourcing to Living Repositories

One of the most innovative aspects is the Community-Based Crowdsourcing module. Instead of a top-down academic approach, the portal allows community members (informants) to upload digital resources directly, causing the ontology to grow dynamically as different villages contribute their unique variations of the kulkul tradition.

Experiments & Results: Is it Usable?

The authors conducted a rigorous evaluation using 68 participants from various Balinese universities. They focused on two core metrics from the Technology Acceptance Model (TAM): Perceived Usefulness (PU) and Perceived Ease of Use (PE).

Figure 2: The logic flow for semantic searching tasks.

Key Findings:

• High Task Accuracy: Despite the complexity of semantic searching (which differs significantly from Google-style keyword searches), roughly 70% of users were able to provide fully correct answers to complex cultural queries.
• Positive Sentiment: The Mean score for Usefulness was 5.95/7, indicating that participants saw real value in the portal for learning and documenting their history.
• Community Desire: Interestingly, several participants proactively asked for access to populate missing details for their own villages, proving the "living repository" concept is socially viable.

Critical Insight: The "Living" Archive

The true success of this paper isn't just the software—it's the Identity Re-affirmation. By providing a structured way for the Balinese community to digitize their own traditions, the authors have solved a major hurdle in heritage studies: the "Extinction" of knowledge.

However, a potential limitation is the Semantic Gap; as rituals evolve, the ontology itself may need periodic updates to reflect modern changes in Balinese society. Future work could benefit from integrating AI-driven automated tag suggestions to help community members categorize their uploads more accurately.

Conclusion

This study bridges the gap between ancient Balinese tradition and modern Semantic Web technologies. It provides a blueprint for how other cultures can utilize cloud computing and ontologies to ensure their voices aren't just recorded, but are organized and accessible for centuries to come.