newsroom

NEWSROOM is a large dataset for training and evaluating summarization systems.
It contains 1.3 million articles and summaries written by authors and
editors in the newsrooms of 38 major publications.

Dataset features includes:
- text: Input news text.
- summary: Summary for the news.
And additional features:
- title: news title.
- url: url of the news.
- date: date of the article.
- density: extractive density.
- coverage: extractive coverage.
- compression: compression ratio.
- density_bin: low, medium, high.
- coverage_bin: extractive, abstractive.
- compression_bin: low, medium, high.

This dataset can be downloaded upon requests. Unzip all the contents
"train.jsonl, dev.josnl, test.jsonl" to the tfds folder.

NEWSROOM 是一项用于训练与评估摘要系统的大型数据集。该数据集汇聚了来自38家主要新闻机构新闻编辑室中的作者和编辑撰写的130万篇新闻文章及其摘要。

数据集特性包括：
- text：输入的新闻文本。
- summary：新闻摘要。

此外，还包含以下附加特性：
- title：新闻标题。
- url：新闻的URL链接。
- date：文章的日期。
- density：提取密度。
- coverage：提取覆盖率。
- compression：压缩比率。
- density_bin：低、中、高。
- coverage_bin：提取式、抽象式。
- compression_bin：低、中、高。

本数据集可应请求下载。将“train.l”、“dev.l”和“test.l”所有内容解压至tfds文件夹。