README.md

metadata

license: cc-by-sa-4.0
task_categories:
  - text-generation
  - text-classification
language:
  - en
tags:
  - reddit
  - social-media
  - scored.co
  - web-scrape
  - parquet
size_categories:
  - 10M<n<100M
pretty_name: Scored.co 2025 scrape

Scored.co 2025 scrape

This dataset is a full scrape of public content from Scored.co, a right-wing Reddit-style social media site. It contains 73,045,361 rows of posts and comments, stored as parquet.

The dataset is intended for research into online communities, political discussion, social media moderation, misinformation, platform migration, network dynamics, and large-scale text analysis.

Data format

The main dataset is partitioned by entity type:

• entity_type=post/: post records
• entity_type=comment/: comment records

Hugging Face also provides converted parquet shards on the refs/convert/parquet branch, which can be useful for lightweight inspection or streaming workflows without downloading the full dataset.

Example:

from datasets import load_dataset

# Streams records without downloading the full dataset first.
ds = load_dataset(
    "trentmkelly/scored_co_2025",
    split="train",
    streaming=True,
)

for row in ds.take(3):
    print(row)

You can also work directly with parquet URLs or the hf:// filesystem interface if using tools such as DuckDB, Polars, PyArrow, or pandas.

Columns

A sample shard from refs/convert/parquet had the following fields:

Column	Description
post_id	Numeric Scored.co post ID.
post_uuid	Scored.co post UUID/string identifier.
comment_id	Numeric comment ID, when the row is a comment.
comment_uuid	Comment UUID/string identifier, when available.
parent_id	Parent post or comment identifier.
author	Public username associated with the record.
community	Scored.co community name.
title	Post title. For comments, this may repeat the parent post title.
body	Plain-text body/content extracted from the record.
created_ms	Creation timestamp in Unix milliseconds.
created_iso	Creation timestamp as an ISO-8601 UTC string.
score	Total score.
score_up	Upvote score/count, when available.
score_down	Downvote score/count, when available.
comments_total	Total comment count for post rows, when available.
top_level_comments	Top-level comment count for post rows, when available.
depth	Comment nesting depth, when available.
is_deleted	Whether the item was marked deleted.
is_removed	Whether the item was marked removed.
is_nsfw	Whether the item was marked NSFW.
is_image	Whether the item is an image post.
is_video	Whether the item is a video post.
vote_state	Vote-state value from the source record.
pro_tier	Source account/profile tier value.
link	Linked URL for link posts, when present.
domain	Domain extracted from link, when present.
source_file	Original scrape/source JSON filename.
raw_json	Raw source JSON payload retained as a string.

Some fields are entity-specific and may be null for posts or comments.

Notes and limitations

• This is scraped public web data and may include offensive, hateful, explicit, false, or otherwise sensitive content.
• Usernames and other public identifiers are included. Handle the data carefully and avoid attempts to identify, contact, harass, or profile individual users.
• Text may contain HTML-derived artifacts, links, markdown, quoted material, deleted/removed content markers, or source-platform metadata.
• The dataset reflects what was available to the scraper at collection time; it should not be treated as a complete or current representation of Scored.co.

Citation

If you use this dataset, cite the Hugging Face dataset repository:

@dataset{kelly_scored_co_2025,
  author = {Kelly, Trent M.},
  title = {Scored.co 2025 scrape},
  year = {2025},
  publisher = {Hugging Face},
  url = {https://huggingface.co/datasets/trentmkelly/scored_co_2025}
}