Giters

hsc748NLP / SikuBERT-for-digital-humanities-and-classical-Chinese-information-processing

SikuBERT：四库全书的预训练语言模型（四库BERT） Pre-training Model of Siku Quanshu

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool



GitHub license

[TOC]

中文版

引言

数字人文研究需要大规模语料库和高性能古文自然语言处理工具支持。预训练语言模型已经在英语和现代汉语文本上极大的提升了文本挖掘的精度，目前亟需专门面向古文自动处理领域的预训练模型。

我们以校验后的高质量《四库全书》全文语料作为训练集，基于BERT深度语言模型框架，构建了面向古文智能处理任务的SikuBERT和SikuRoBERTa预训练语言模型。

我们设计了面向《左传》语料的古文自动分词、断句标点、词性标注和命名实体识别4个下游任务，验证模型性能。

  • SikuBERT和SikuRoBERTa基于《四库全书》语料训练，《四库全书》又称《钦定四库全书》，是清代乾隆时期编修的大型丛书。实验去除了原本中的注释部分，仅纳入正文部分，参与实验的训练集共纳入字数达536,097,588个，数据集内的汉字均为繁体中文。

  • 基于领域适应训练（Domain-Adaptive Pretraining）的**，SikuBERT和SikuRoBERTa在BERT结构的基础上结合大量古文语料，分别继续训练BERT和RoBERTa模型，以获取面向古文自动处理领域的预训练模型。

新闻

面向数字人文的古籍智能处理平台sikuaip1.0版本已正式发布

下载链接:https://pan.baidu.com/s/1--S-qyUedIvhBKwapQjPsA 提取码：m36d

平台的使用方法见“使用方法”文件夹，目前版本支持分词，断句，实体识别，文本分类，词性标注和自动标点六种功能，提供单文本处理和语料库处理两种文本处理模式，欢迎下载使用！

古文生成与古诗词生成预训练模型

古文生成预训练模型 SikuGPT2：https://huggingface.co/JeffreyLau/SikuGPT2

古诗词生成预训练模型 SikuGPT2-poem：https://huggingface.co/JeffreyLau/SikuGPT2-poem

论文：AIGC助力数字人文研究的实践探索：SikuGPT驱动的古诗词生成研究：https://kns.cnki.net/kcms/detail/11.1762.G3.20230426.1046.002.html

使用方法

Huggingface Transformers

基于Huggingface Transformers的from_pretrained方法可以直接在线获取SikuBERT和SikuRoBERTa模型。

  • SikuBERT
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("SIKU-BERT/sikubert")

model = AutoModel.from_pretrained("SIKU-BERT/sikubert")
  • SikuRoBERTa
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("SIKU-BERT/sikuroberta")

model = AutoModel.from_pretrained("SIKU-BERT/sikuroberta")

下载模型

  • 我们提供的模型是PyTorch版本。

调用

云盘

旧版下载地址:

模型名称 网盘链接
sikubert 链接 提取码: jn94
sikuroberta 链接 提取码: ihgq

拥有新词表的sikubert和sikuroberta下载方式已更新:

模型名称 网盘链接
sikubert_vocabtxt(推荐下载) 链接 提取码: v68d
sikuroberta_vocabtxt(推荐下载) 链接 提取码: 93cr

验证与结果

  • 我们SikuBERT和SikuRoBERTa应用在语源为繁体中文的《左传》语料上进行古文自动分词、词性标注、断句和实体识别实验。实验结果如下。
任务名 task type 预训练模型pretrained models 精确率（P） 召回率（R） 调和平均值（F1）
分词 Participle BERT-base-chinese 86.99% 88.15% 87.56%
RoBERTa 80.90% 84.77% 82.79%
SikuBERT 88.62% 89.08% 88.84%
SikuRoBERTa 88.48% 89.03% 88.88%
词性标注 POS tag BERT-base-chinese 89.51% 90.10% 89.73%
RoBERTa 86.70% 88.45% 87.50%
SikuBERT 89.89% 90.41% 90.10%
SikuRoBERTa 89.74% 90.49% 90.06%
断句 Segmentation BERT-base-chinese 78.77% 78.63% 78.70%
RoBERTa 66.71% 66.38% 66.54%
SikuBERT 87.38% 87.68% 87.53%
SikuRoBERTa 86.81% 87.02% 86.91%
任务名 task type 预训练模型pretrained models 实体名 entity names 精确率（P） 召回率（R） 调和平均值（F1）
实体识别 NER BERT-base-chinese nr(人名) 86.66% 87.35% 87.00%
ns(地名) 83.99% 87.00% 85.47%
t(时间) 96.96% 95.15% 96.05%
avg/prf 86.99% 88.15% 87.56%
RoBERTa nr(人名) 79.88% 83.69% 81.74%
ns(地名) 78.86% 84.08% 81.39%
t(时间) 91.45% 91.79% 91.62%
avg/prf 80.90% 84.77% 82.79%
SikuBERT nr(人名) 88.65% 88.23% 88.44%
ns(地名) 85.48% 88.20% 86.81%
t(时间) 97.34% 95.52% 96.42%
avg/prf 88.62% 89.08% 88.84%
SikuRoBERTa nr(人名) 87.74% 88.23% 87.98%
ns(地名) 86.55% 88.73% 87.62%
t(时间) 97.35% 95.90% 96.62%
avg/prf 88.48% 89.30% 88.88%

引用

  • 如果我们的内容有助您研究工作，欢迎在论文中引用。

  • GB/T 7714-2015格式： [1]王东波, 刘畅, 朱子赫, 等. SikuBERT与SikuRoBERTa：面向数字人文的《四库全书》预训练模型构建及应用研究[J]. 图书馆论坛, 2022, 42(06): 31-43.

    [2]刘江峰, 刘雏菲, 齐月, 等. AIGC助力数字人文研究的实践探索：SikuGPT驱动的古诗词生成研究[J/OL]. 情报理论与实践: 1-12[2023-04-27]. http://kns.cnki.net/kcms/detail/11.1762.G3.20230426.1046.002.html.

免责声明

  • 报告中所呈现的实验结果仅表明在特定数据集和超参组合下的表现，并不能代表各个模型的本质。实验结果可能因随机数种子，计算设备而发生改变。使用者可以在许可证范围内任意使用该模型，但我们不对因使用该项目内容造成的直接或间接损失负责。

致谢

联系我们

English version

Introduction

Digital humanities research needs the support of large-scale corpus and high-performance ancient Chinese natural language processing tools. The pre-training language model has greatly improved the accuracy of text mining in English and modern Chinese texts. At present, there is an urgent need for a pre-training model specifically for the automatic processing of ancient texts.

We used the verified high-quality "Siku Quanshu" full text corpus as the training set, and based on the BERT deep language model framework, we constructed SikuBERT and SikuRoBERTa pre-training language models for intelligent processing tasks of ancient Chinese.

We designed four downstream tasks of automatic word segmentation, segmentation punctuation, part-of-speech tagging, and named entity recognition for ancient Chinese corpus for "Zuo Zhuan" to verify the performance of the model.

  • SikuBERT and SikuRoBERTa are trained on the corpus of "Siku Quanshu". "Siku Quanshu", also known as "King Ding Siku Quanshu", is a large-scale series of books compiled during the Qianlong period of the Qing Dynasty. The experiment removed the original annotation part and only included the text part. The training set involved in the experiment included a total of 536,097,588 characters, and the Chinese characters in the data set were all traditional Chinese.
  • Based on the idea of Domain-Adaptive Pretraining, SikuBERT and SikuRoBERTa combine a large amount of ancient text corpus based on the BERT structure, and continue to train the BERT and RoBERTa models respectively to obtain pre-training models for the automatic processing of ancient texts.

News

  • 2021/5/8 model added Huggingface Transformers pre-trained model "Family Bucket".
  • 2021/5/6 papers accepted by the 5th National Future Smart Library Development Forum
  • 2021/8/20 papers will be published in "Library Forum" [Internet First](https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CAPJ&dbname=CAPJLAST&filename=TSGL20210819003&v=hVxys9Ams5mx5uSYJEe%25mmd2BJZlCnoc0g2INTV% 25mmd2F8yn71Wrz56s)
  • 2021/9/13 update sikuBERT and sikuRoberta, the newly released model has a new vocabulary that contains the original words of "Siku Quanshu", the new vocabulary has more than 8,000 words more than the original bert-base vocabulary. The performance on all tasks surpassed the former.
  • 2021/9/15 related python toolkit sikufenci is officially released, which can be used for automatic word segmentation of traditional ancient books, see https://github.com/SIKU-BERT/sikufenci for the link
  • 2021/11/6 The stand-alone open source software sikuaip related to this project is officially released, providing various ancient text processing functions including word segmentation, sentence segmentation, entity recognition, text classification, etc., which can be downloaded and decompressed directly. See the link below.
  • 2021/12/10 The first NLP tool evaluation competition in the field of ancient Chinese based on this model-EvaHan 2022 is released. For details of the competition, please visit: https://circse.github.io/LT4HALA/2022/EvaHan
  • 2022 The generative pre-training model of ancient Chinese and ancient Poetry based on the SiKu QuanShu and Chinese-poetry is published, as detailed in details：https://huggingface.co/JeffreyLau/SikuGPT2 https://huggingface.co/JeffreyLau/SikuGPT2-poem
  • 2022/10/21 The TfhBERT, a classical text pretraining language model based on the Twenty-Four Histories, has been released. For more details, please refer to: https://huggingface.co/hsc748NLP/TfhBERT. The BTfhBER, a cross-lingual pretraining model for ancient texts, has been released. For more details, please refer to: https://huggingface.co/hsc748NLP/BtfhBERT.
  • 2023/7 The "Guji" series models were officially released at the 2023 China Intelligence Conference and the Bitren(比特人文) WeChat Official Account. The new version of the "Guji" series models includes a total of 9 pretraining models in 3 categories, catering to the different processing preferences of researchers in ancient text studies. For the download links of the "Guji" series models, please refer to: https://github.com/hsc748NLP/GujiBERT-and-GujiGPT, https://huggingface.co/hsc748NLP.

The sikuaip version 1.0 of the ancient book intelligent processing platform for digital humanities has been officially released

Download link: https://pan.baidu.com/s/1--S-qyUedIvhBKwapQjPsA Extraction code: m36d

Please refer to the "How to use" folder for the usage of the platform. The current version supports six functions of word segmentation, sentence segmentation, entity recognition, text classification, part-of-speech tagging and automatic punctuation. It provides two text processing modes: single text processing and corpus processing. Welcome to download and use !

Pre-training model of ancient Chinese text generation and ancient Chinese poetry generation

SikuGPT2: https://huggingface.co/JeffreyLau/SikuGPT2

SikuGPT2-poem: https://huggingface.co/JeffreyLau/SikuGPT2-poem

A Practical Exploration ofAIGC-Powered Digital Humanities Research: A SikuGPT Driven Research of Ancient Poetry Generation：https://kns.cnki.net/kcms/detail/11.1762.G3.20230426.1046.002.html

How to use

Huggingface Transformers

The from_pretrained method based on Huggingface Transformers can directly obtain SikuBERT and SikuRoBERTa models online.

  • SikuBERT
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("SIKU-BERT/sikubert")

model = AutoModel.from_pretrained("SIKU-BERT/sikubert")
  • SikuRoBERTa
from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained("SIKU-BERT/sikuroberta")

model = AutoModel.from_pretrained("SIKU-BERT/sikuroberta")

Download PTM

  • The model we provide is the PyTorch version.

From Huggingface

From Google Drive

If you are not in China, We put the model on Google drive for users to downloa.

Old version download address:

Model Link
sikubert https://drive.google.com/drive/folders/1blElNRhouuaU-ZGA99ahud1QL7Y-7PEZ?usp=sharing
sikuroberta https://drive.google.com/drive/folders/13ToN58XfsfHIIj7pjLWNgLvWAqHsb0_a?usp=sharing

The download method of sikubert and sikuroberta with new vocabulary has been updated:

Model Link
sikubert_vocabtxt(Recommended download ) https://drive.google.com/drive/folders/1uA7m54Cz7ZhNGxFM_DsQTpElb9Ns77R5?usp=sharing
sikuroberta_vocabtxt(Recommended download ) https://drive.google.com/drive/folders/1i0ldNODE1NC25Wzv0r7v1Thda8NscK3e?usp=sharing

Evaluation & Results

  • We use SikuBERT and SikuRoBERTa to perform automatic word segmentation, part-of-speech tagging, sentence segmentation and entity recognition experiments on the "Zuo Zhuan" corpus whose etymology is traditional Chinese. The experimental results are as follows.
task type pretrained models （P） （R） （F1）
Participle BERT-base-chinese 86.99% 88.15% 87.56%
RoBERTa 80.90% 84.77% 82.79%
SikuBERT 88.62% 89.08% 88.84%
SikuRoBERTa 88.48% 89.03% 88.88%
POS tag BERT-base-chinese 89.51% 90.10% 89.73%
RoBERTa 86.70% 88.45% 87.50%
SikuBERT 89.89% 90.41% 90.10%
SikuRoBERTa 89.74% 90.49% 90.06%
Segmentation BERT-base-chinese 78.77% 78.63% 78.70%
RoBERTa 66.71% 66.38% 66.54%
SikuBERT 87.38% 87.68% 87.53%
SikuRoBERTa 86.81% 87.02% 86.91%
task type pretrained models entity names （P） （R） （F1）
NER BERT-base-chinese nr(people name) 86.66% 87.35% 87.00%
ns(place name) 83.99% 87.00% 85.47%
t(time) 96.96% 95.15% 96.05%
avg/prf 86.99% 88.15% 87.56%
RoBERTa nr(people name) 79.88% 83.69% 81.74%
ns(place name) 78.86% 84.08% 81.39%
t(time) 91.45% 91.79% 91.62%
avg/prf 80.90% 84.77% 82.79%
SikuBERT nr(people name) 88.65% 88.23% 88.44%
ns(place name) 85.48% 88.20% 86.81%
t(time) 97.34% 95.52% 96.42%
avg/prf 88.62% 89.08% 88.84%
SikuRoBERTa nr(people name) 87.74% 88.23% 87.98%
ns(place name) 86.55% 88.73% 87.62%
t(time) 97.35% 95.90% 96.62%
avg/prf 88.48% 89.30% 88.88%

Citing

  • If our content is helpful for your research work, please quote it in the paper.

  • GB/T 7714-2015格式： [1]王东波, 刘畅, 朱子赫, 等. SikuBERT与SikuRoBERTa：面向数字人文的《四库全书》预训练模型构建及应用研究[J]. 图书馆论坛, 2022, 42(06): 31-43.

    [2]刘江峰, 刘雏菲, 齐月, 等. AIGC助力数字人文研究的实践探索：SikuGPT驱动的古诗词生成研究[J/OL]. 情报理论与实践: 1-12[2023-04-27]. http://kns.cnki.net/kcms/detail/11.1762.G3.20230426.1046.002.html.

    [3] Wang D, Liu C, Zhao Z, et al. GujiBERT and GujiGPT: Construction of Intelligent Information Processing Foundation Language Models for Ancient Texts[J]. arXiv preprint arXiv:2307.05354, 2023.

Disclaim

  • The experimental results presented in the report only show the performance under a specific data set and hyperparameter combination, and cannot represent the essence of each model. The experimental results may change due to random number seeds and computing equipment. **Users can use the model arbitrarily within the scope of the license, but we are not responsible for the direct or indirect losses caused by using the content of the project. **

Acknowledgment

Contact us

About

SikuBERT：四库全书的预训练语言模型（四库BERT） Pre-training Model of Siku Quanshu

License:Apache License 2.0