---
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. |

##### 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.