HADIA KB as an Evidence Preserving Digital Knowledge Infrastructure for Traditional Rice Beer in Jharkhand India

Abstract Traditional knowledge about Handia, Hadia and Haria is valuable but digitally fragmented across ethnobotany, fermentation science, microbiology, nutrition, cultural studies, policy and law. We present HADIA-KB, an evidence-linked knowledge infrastructure that preserves source provenance, reported terminology and geographic scope instead of collapsing heterogeneous regional traditions into a single recipe or truth claim. Version 1.0 contains 29 indexed sources, 34 verified claim records, 16 plant or lichen records, 12 microorganisms, 18 quantitative measurements and 192 source-carrying graph relations. A reproducible pipeline converts public documentation into normalized entities, claim-level evidence passports, contextual-difference records and machine-readable graph exports. Empirical analysis shows a marked documentation imbalance: only 12 of 34 claims concern Jharkhand, while microbiology, chemistry, nutrition and metabolomics have no Jharkhand-specific claim in the frozen corpus. The evidence graph contains 125 nodes, 192 directed edges and two weakly connected components; claim attribution is distributed across an effective 22.27 sources rather than dominated by a single document. HADIA-KB therefore contributes not another description of rice beer, but an auditable map of what is documented, where evidence applies, where sources differ, and what remains unstudied. The accompanying software supports exploration, filtered export, graph analysis and reproducible extension by researchers, communities, and policymakers worldwide.