Datasets:
File size: 3,378 Bytes
b36864d ffcedfa b36864d | 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 | ---
pretty_name: Causals Graph
license: cc-by-4.0
task_categories:
- graph-ml
tags:
- knowledge-graph
- news
- events
- causality
- graph
configs:
- config_name: nodes
data_files:
- split: snapshot
path: v0.1.0-2026-10-09/nodes.jsonl.gz
- config_name: edges
data_files:
- split: snapshot
path: v0.1.0-2026-10-09/edges.jsonl.gz
---
# Causals Graph
A versioned, news-derived graph snapshot from **Causals Lab**. Version
`v0.1.0-2026-10-09` contains 369,184 nodes and 1,272,418 directed edges.
The graph contains descriptive and contextual relationships as well as
extracted causal *claims*. It is **not** a collection of independently
verified causal effects; a graph path is not automatically a causal path.
This Hugging Face copy mirrors the [canonical GitHub release](https://github.com/causals-lab/community/tree/graph-v0.1.0-2026-10-09/data/v0.1.0-2026-10-09).
The [Causals Lab download page](https://causals.org/download/) gives an
overview. Use the snapshot version and SHA-256 hashes in `manifest.json` to
identify the exact data. The two Viewer configurations, `nodes` and `edges`,
are tables from the same graph snapshot; join an edge's `source_id` and
`target_id` to node `id`.
## Get started in 10 minutes
Try the [runnable notebook](https://github.com/causals-lab/community/blob/main/quickstart/quickstart.ipynb) or [open it in Colab](https://colab.research.google.com/github/causals-lab/community/blob/main/quickstart/quickstart.ipynb). It uses a [small downloadable sample](https://github.com/causals-lab/community/tree/main/quickstart/sample) of 62 nodes and 9 edges selected from this exact release. It checks file hashes, joins edges to nodes, and inspects source-news IDs without requiring private code, a database, or an API key. The selected sample is not representative and does not establish causal effects.
## Files
The folder `v0.1.0-2026-10-09/` contains:
- `nodes.jsonl.gz` — 369,184 UTF-8 JSONL records, gzip-compressed.
- `edges.jsonl.gz` — 1,272,418 UTF-8 JSONL records, gzip-compressed.
- `manifest.json` — counts, relation distribution, file sizes and SHA-256 hashes.
- `DATA_CARD.md` — schema, provenance, limitations and quality cautions.
- `LICENSE_DATA.md` — dataset reuse terms.
Read [the complete data card](./v0.1.0-2026-10-09/DATA_CARD.md) before
interpreting relations or publishing analysis. In particular, `MENTIONS` is
not causal; `OTHER` is unclassified; and `TRIGGERED` or `STATED_CAUSE` are
extracted or attributed claims, not independent intervention evidence.
The snapshot has no representative whole-graph causal-validity estimate.
## License, attribution and corrections
The released dataset is offered under
[CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Attribute
**Causals Lab**, identify the version, link to the source and license, and
indicate changes. This license applies only to rights the publisher can grant
in the released files; it does not license third-party news content, private
software, trademarks, or privacy/publicity rights.
Suggested citation:
> Causals Lab. (2026). *Causals Graph* (v0.1.0-2026-10-09) [Data set].
> https://github.com/causals-lab/community/tree/graph-v0.1.0-2026-10-09/data/v0.1.0-2026-10-09
For data corrections or removal concerns, write privately to
**data@causals.org**. Do not post personal or sensitive details in a public
issue.
|