Dataset Card for MNBVC
Table of Contents
Dataset Description
- Homepage: http://mnbvc.253874.net/
- Repository: https://github.com/esbatmop/MNBVC
- Paper: N/A
- Leaderboard: N/A
- Point of Contact: N/A
数据集介绍
中文互联网上最古老最神秘(没有之一)的MOP里屋社区于2023.1.1庄重宣布:
在英明神武的猫扑管子带领下,决心发挥社区所长(哪都长),帮助开源社区长期更新一份最大的中文互联网语料集。
Huggingface上的MNBVC数据集在逐渐更新中,请到https://github.com/esbatmop/MNBVC 获取未完成清洗的更多数据。
可以使用如下脚本加载:
from datasets import load_dataset
# 对应语料数据加载
# 请参考: 下面表格中的标签字段字段内容
# 如:序号1,arXiv文献的文本。,标签字典:academic_paper
dataset_arxiv = load_dataset("liwu/MNBVC", 'academic_paper', split='train', streaming=True)
# 如:序号38,法律判决书文本,标签字典:law_judgement
dataset_law_judgement = load_dataset("liwu/MNBVC", 'law_judgement', split='train', streaming=True)
next(iter(dataset)) # get the first line
数据子集
MNBVC数据集包含数个子集:
| 序号 | 一级目录 | 二级目录 | 描述说明 | 标签字典 | 备注 |
|---|---|---|---|---|---|
| 1 | academic_paper | - | 来自文献的文本。 | - | - |
| 1.1 | academic_paper | arxiv | 来自arXiv文献的文本。 | academic_paper | - |
| 2 | blog | - | 博客语料目录 | blog | - |
| 2.1 | blog | 163_blog | - | blog | - |
| 2.2 | blog | ai_blog | - | blog | - |
| 2.3 | blog | it_blog | - | blog | - |
| 3 | book | - | 书籍语料目录 | book | - |
| 3.1 | book | InfoSec | - | book | - |
| 4 | co_ann_report | - | 企业年报文本。 | co_ann_report | - |
| 4.1 | code | - | 来自代码的文本 | - | - |
| 4.2 | code | metadata | github 仓库的代码元数据 | code | - |
| 4.3 | code | googlecode | github 中,不同仓库的数据 | code | - |
| 4.4 | code | githubcode | google code 中, 不同仓库的数据 | code | |
| 5 | crawler | - | 爬虫语料目录 | - | |
| 5.1 | crawler | oscar | 从CommonCrawl中清洗出来的通用文本数据。 | crawler_oscar | - |
| 6 | forum | - | 论坛语料目录 | forum | - |
| 7 | game | - | 一些游戏的平行语料数据。 | - | |
| 7.1 | game | Baldurs_Gate_3 | 博德之门 3 | game | - |
| 7.2 | game | DarkSouls3 | 黑暗之魂III | game | - |
| 7.3 | game | do_not_starve | 饥荒 | game | - |
| 7.4 | game | EldenRing | 艾尔登法环 | game | - |
| 7.5 | game | Genshin_Anime | 原神 | game | - |
| 7.6 | game | GTA | 侠盗猎车手4 与 侠盗猎车手5 | game | - |
| 7.7 | game | Hogwarts_legacy | 霍格沃茨指遗 | game | - |
| 7.8 | game | hades | 哈迪斯 | game | - |
| 7.9 | game | Ib | Ib恐怖美术馆 | game | - |
| 7.10 | game | RDR2RE | 碧血狂殺2 | game | - |
| 7.11 | game | sekiro | 只狼 | game | - |
| 7.12 | game | Sid_Meiers_CivilizationVI | 文明VI | game | - |
| 7.13 | game | slay_the_spire | 杀戮尖塔 | game | - |
| 7.14 | game | StarRail | 崩坏:星穹铁道 | game | - |
| 7.15 | game | stellaris | 群星 | game | - |
| 7.16 | game | Terraria | 泰拉瑞亚 | game | - |
| 7.17 | game | The_Wither_3 | 巫师三 | game | - |
| 7.18 | game | Turing_Complete | 图灵完备性 | game | - |
| 7.19 | game | witchspring | 魔女之泉R | game | - |
| 7.20 | game | Wuthering | 鸣潮 | game | - |
| 7.21 | game | Yakuza | 人中之龙 | game | - |
| 8 | gov | - | 政府资料目录 | - | |
| 8.1 | gov | xuexiqiangguo | 来自学习强国的文本。 | gov_xuexiqiangguo | - |
| 8.2 | gov | gov_report | 来自政府工作报告的文本。 | gov_report | - |
| 9 | law | - | 来自法律文书的文本。 | - | |
| 9.1 | law | judgement | 法律判决书文本。 | law_judgement | - |
| 10 | math | - | 与数学相关的中文语料 | math | - |
| 10.1 | math | qa | 和数学领域有关的问答数据。 | math_qa | - |
| 10.2 | math | emath | 中国数学爱好者论坛语料数据 | emath | - |
| 10.3 | math | chat | 和数学领域有关的对话数据数据,可以提升模型Chain of Thought的能力。 | math_chat | - |
| 11 | new | - | 来自新闻的的文本数据 | new | - |
| 11.1 | news | peoples_daily | 来自人民日报的文本数据。 | new | - |
| 12 | parallel | - | 平行语料目录 | - | - |
| 12.1 | parallel | subtitle | 字幕语料 | - | - |
| 12.1.1 | parallel | subtitle \ yyets | 人人影视 | parallel_subtitle_yyets | - |
| 12.1.2 | parallel | subtitle \ shooter.cn | 射手网 | parallel_subtitle_shooter | - |
| 12.2 | parallel | united_nations | 联合国平行语料 | parallel_united_nations | - |
| 13 | patent | - | 专利文本数据目录 | - | |
| 14 | qa | - | 来自各大问答语料 | - | |
| 14.1 | qa | chatgpt | 使用ChatGPT构造的问答语料,感谢genggui001贡献语料。 | qa_chatgpt | - |
| 14.2 | qa | mfa | 外交部问答数据。 | qa_mfa | - |
| 14.3 | qa | quora | 来自quora 网站的问答语料 | qa_quora | - |
| 14.4 | qa | stackexchange | 来自StackExchange的问答数据。 | qa_stackexchange | - |
| 14.5 | qa | wikihow | 来自wikihow的问答数据。 | qa_wikihow | - |
| 14.6 | qa | zhihu | 来自知乎的问答数据。 | qa_zhihu | - |
| 15 | wikipedia | - | 来自维基百科的文本数据。 | wikipedia | - |
| 16 | medical | - | 来自药品说明书 | medical | - |
数据格式
目前MNBVC数据集包含如下几类数据:
- 通用文本
- 问答语料
- 代码语料
- 多轮对话
- 论坛语料
- 平行语料
可以在MNBVC的wiki页面上查看这几类数据的具体格式。
项目早期所上传的数据使用如下格式,以后这一格式会被废弃,相应数据也会重新上传:
{
"text": datasets.Value("string"),
"meta": datasets.Value("string")
}
Contributions
Thanks to the Liwu community for constructing this dataset. Thanks to silver 、 jiaming and Mark Leung for adding and uploading this dataset to Huggingface.
Citation
Please cite the repo if you use the data or code in this repo.
@misc{mnbvc,
author = {{MOP-LIWU Community} and {MNBVC Team}},
title = {MNBVC: Massive Never-ending BT Vast Chinese corpus},
year = {2023},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/esbatmop/MNBVC}},
}