CoolFace
Datasetpublic

justicedao/ipfs_netherlands_laws_vector_index

IPFS Netherlands Laws Vector Index Hugging Face target: justicedao/ipfs_netherlands_laws_vector_index. Dense vector mapping keyed by source CID, with FAISS and TF-IDF/SVD artifacts. This index covers 94736 rows from the paired CID dataset. Source scrape date: 2026-06-27T14:04:23.964996. Full BWB discovery inventory found 42,956 unique BWBR identifiers from official SRU discovery; this paired snapshot contains 5,000 completed identifiers and must not be described as the full… See the full description on the dataset page: https://huggingface.co/datasets/justicedao/ipfs_netherlands_laws_vector_index.

sourceHugging Faceotherupdated 3mo agoView on Hugging Face
0likes14downloads
Dataset Card

IPFS Netherlands Laws Vector Index

Hugging Face target: justicedao/ipfs_netherlands_laws_vector_index.

Dense vector mapping keyed by source CID, with FAISS and TF-IDF/SVD artifacts.

This index covers 94736 rows from the paired CID dataset. Source scrape date: 2026-06-27T14:04:23.964996. Full BWB discovery inventory found 42,956 unique BWBR identifiers from official SRU discovery; this paired snapshot contains 5,000 completed identifiers and must not be described as the full Dutch corpus. The paired base dataset includes article extraction diagnostics and parser coverage improvements for older/French heading styles. The paired source package was quality-audited after parser-noise cleanup for official website UI chrome, duplicate article identifier disambiguation, clean hierarchy/citation validation, and sampled CID/vector/BM25/KG retrieval checks. Historical/former laws are preserved and labeled with law_status, is_current, validity dates, status source, confidence, and notes; these fields are also present in the vector mapping rows for filtering. This index is not legal advice.