Datasets:
File size: 5,542 Bytes
ec63fd7 c05cffc ec63fd7 c05cffc ec63fd7 c05cffc ec63fd7 df3aae3 c05cffc ec63fd7 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 | ---
license: cc-by-sa-4.0
language:
- en
pretty_name: wikiparse 2606
tags:
- wikipedia
- citations
- bibliometrics
- data-quality
configs:
- config_name: refs
data_files:
- split: train
path: data/refs/0000.parquet
- config_name: articles
data_files:
- split: train
path: data/articles/0000.parquet
- config_name: categories
data_files:
- split: train
path: data/categories/0000.parquet
- config_name: publishers
data_files:
- split: train
path: data/publishers/0000.parquet
- config_name: templates
data_files:
- split: train
path: data/templates/0000.parquet
- config_name: clusters
data_files:
- split: train
path: data/clusters/0000.parquet
- config_name: streams
data_files:
- split: train
path: data/streams/0000.parquet
- config_name: disambig-nodes
data_files:
- split: train
path: data/disambig-nodes/0000.parquet
- config_name: disambig-links
data_files:
- split: train
path: data/disambig-links/0000.parquet
- config_name: redirects
data_files:
- split: train
path: data/redirects/0000.parquet
---
# wikiparse 2606
Ten Parquet tables derived from the English Wikipedia 2026-06-01 dump.
The useful unit here is a table, not a monolithic corpus download: each config
has one grain, one schema, one checksum, and one job it is reasonably good at.
##### Tables
| Config | Rows | What it is good for |
| ---------------- | ---------: | ---------------------------------------------------------------- |
| `refs` | 256,564 | Raw ref and cite syntax counts by multistream member. |
| `articles` | 5,135,607 | Article-level citation counts and parser-quality signals. |
| `categories` | 529,402 | Category-level article and citation aggregates. |
| `publishers` | 20,000 | A bounded publisher-string projection with quality coordinates. |
| `templates` | 37 | Citation-template families by parser outcome. |
| `clusters` | 1,511,504 | Exact-signature citation clusters for deterministic repair work. |
| `streams` | 256,564 | Per-stream custody and citation-spine coverage signals. |
| `disambig-nodes` | 310,529 | Typed disambiguation nodes with explicit relation status. |
| `disambig-links` | 723,486 | Heuristic disambiguation-title candidate links. |
| `redirects` | 11,894,129 | Redirect edges with target-existence and disambiguation flags. |
##### Context delivery
The public [`wikiparse-2606-data` bucket](https://huggingface.co/buckets/robbd/wikiparse-2606-data)
holds the larger, mutable delivery surface. The dataset repo remains the smaller, versioned
catalog.
| Priority | Articles | Intended use |
| -------- | --------: | --------------------------------------------------------------------------------------- |
| `p0` | 908,270 | Science closure and taxa; first-pass research context. |
| `p1` | 4,665,480 | General retained articles. |
| `p2` | 1,208,553 | Useful secondary context such as settlements, films, music releases, and sports people. |
| `p3` | 291,274 | Serial/incidental context such as lists, sports seasons, and disambiguation pages. |
These four labels are mutually exclusive delivery hints, not quality judgments or exclusion
authority. The index preserves categories and detector signals so consumers can write a better
policy. Article-content shards are copied from the existing compact corpus; they do not rerun the
parser.
Content `v1` has 7,073,577 rows in 1,003 disjoint shards (18,248,540,799 payload bytes).
`context/content/v1/manifest.json` has SHA-256 `f3de8e5b6c4f0a2d26f5f17d27a915b6964c16c69707a38de24af7d6647569c6`;
the publication receipt has SHA-256 `71bb9b65bab44d6b9237a0b6da400a653f547d420fdcae4e4172a9818f12147e`.
##### Analysis
5 small, source-bound cards summarize the least ambiguous observations:
- [category-structure](analysis/category-structure.md)
- [citation-quality](analysis/citation-quality.md)
- [provenance-repair](analysis/provenance-repair.md)
- [science-closure](analysis/science-closure.md)
- [wave0-landscape](analysis/wave0-landscape.md)
##### What this is not
- Parser signals such as `suspect` are operational labels, not adjudicated truth.
- Category counts overlap. Do not read their sums as distinct population totals.
- Disambiguation links are heuristic candidates; they are not an authoritative taxonomy.
- Citation records are article-level and deduplicated. They are not inline claim markers.
- This is a technical milestone, not a paper, benchmark leaderboard, or announcement.
##### Provenance
Every public payload is byte-identical to a reviewed local object. `provenance.json`
binds the two reviewed source inventories, each public file, its row count, and its
Arrow schema. `checksums.sha256` covers the complete public tree.
Parser and release tooling: https://gitlab.com/doering-ai/apps/wikiparse
##### License
The data is derived from English Wikipedia and is distributed under
[CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/).
See `LICENSE.md` for attribution and modification notes.
|