IndustryCorpus_technology

[[中文主页]](README_ZH.md)

Industry models play a crucial role in driving enterprise intelligence transformation and innovative development. High-quality industry data is key to improving the performance of large models and realizing industry applications. However, datasets currently used for industry model training generally suffer from issues such as insufficient data volume, low quality, and lack of domain expertise.

To address these problems, we constructed and applied 22 industry data processing operators to clean and filter 3.4TB of high-quality multi-industry classified Chinese and English language pre-training datasets from over 100TB of open-source datasets including WuDaoCorpora, BAAI-CCI, redpajama, and SkyPile-150B. The filtered data consists of 1TB of Chinese data and 2.4TB of English data. To facilitate user utilization, we annotated the Chinese data with 12 types of labels including alphanumeric ratio, average line length, language confidence score, maximum line length, and perplexity.

Furthermore, to validate the dataset's performance, we conducted continued pre-training, SFT, and DPO training on a medical industry demonstration model. The results showed a 20% improvement in objective performance and a subjective win rate of 82%.

Industry categories: 18 categories including medical, education, literature, finance, travel, law, sports, automotive, news, etc.
Rule-based filtering: Traditional Chinese conversion, email removal, IP address removal, link removal, Unicode repair, etc.
Chinese data labels: Alphanumeric ratio, average line length, language confidence score, maximum line length, perplexity, toxicity character ratio, etc.
Model-based filtering: Industry classification language model with 80% accuracy
Data deduplication: MinHash document-level deduplication
Data size: 1TB Chinese, 2.4TB English

Industry classification data size:

| Industry Category | Data Size (GB) | Industry Category | Data Size (GB) |
| :-------------------:|:----------------:|:-------------------:|:----------------:|
| Programming | 4.1 | Politics | 326.4 |
| Law | 274.6 | Mathematics | 5.9 |
| Education | 458.1 | Sports | 442 |
| Finance | 197.8 | Literature | 179.3 |
| Computer Science | 46.9 | News | 564.1 |
| Technology | 333.6 | Film & TV | 162.1 |
| Travel | 82.5 | Medicine | 189.4 |
| Agriculture | 41.6 | Automotive | 40.8 |
| Emotion | 31.7 | Artificial Intelligence | 5.6 |
| Total (GB) | 3386.5 | | |

For the convenience of users to download and use, we have split the large dataset into sub-datasets for 18 industries. The current one is the sub-dataset for the technology industry.

Data processing workflow:


[中文主页](README_ZH.md)

产业模型（Industry models）对于推动企业智能化转型与创新发展具有关键支撑作用。高质量的产业数据是提升大语言模型（Large Language Model，LLM）性能、实现产业落地应用的核心要素。但当前用于产业模型训练的数据集普遍存在数据体量不足、质量偏低、缺乏领域专业知识等痛点。

为解决上述问题，我们构建并应用了22种产业数据处理算子，从包含悟道语料库（WuDaoCorpora）、BAAI-CCI、RedPajama、SkyPile-150B在内的超100TB开源数据集当中，清洗筛选出3.4TB高质量多产业分类中英双语预训练数据集；其中中文数据体量为1TB，英文数据体量为2.4TB。为便于用户使用，我们为中文数据标注了12类标签，涵盖字符数字占比、平均行长度、语言置信度得分、最大行长度以及困惑度（perplexity）等指标。

此外，为验证本数据集的性能表现，我们针对医疗产业示范模型开展了持续预训练、监督微调（Supervised Fine-Tuning，SFT）以及直接偏好优化（Direct Preference Optimization，DPO）训练，实验结果显示模型客观性能提升20%，主观胜率达82%。

产业类别涵盖医疗、教育、文学、金融、旅游、法律、体育、汽车、新闻等共18个大类。
基于规则的过滤流程包括繁体转换、邮箱移除、IP地址移除、链接移除、Unicode修复等操作。
中文数据标签包含字符数字占比、平均行长度、语言置信度得分、最大行长度、困惑度、有害字符占比等。
基于模型的过滤流程采用准确率达80%的产业分类大语言模型。
数据去重采用MinHash文档级去重方法。
数据总体量：中文1TB，英文2.4TB。

产业分类数据体量详情：

| 产业类别 | 数据体量（GB） | 产业类别 | 数据体量（GB） |
|:-------------------:|:----------------:|:-------------------:|:----------------:|
| 编程（Programming） | 4.1 | 政治（Politics） | 326.4 |
| 法律（Law） | 274.6 | 数学（Mathematics） | 5.9 |
| 教育（Education） | 458.1 | 体育（Sports） | 442.0 |
| 金融（Finance） | 197.8 | 文学（Literature） | 179.3 |
| 计算机科学（Computer Science） | 46.9 | 新闻（News） | 564.1 |
| 科技（Technology） | 333.6 | 影视（Film & TV） | 162.1 |
| 旅游（Travel） | 82.5 | 医疗（Medicine） | 189.4 |
| 农业（Agriculture） | 41.6 | 汽车（Automotive） | 40.8 |
| 情感（Emotion） | 31.7 | 人工智能（Artificial Intelligence） | 5.6 |
| 总计（GB） | 3386.5 | | |

为方便用户下载与使用，我们将总数据集拆分为18个产业的子数据集，当前数据集即为科技产业子数据集。

数据处理工作流：