Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,51 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
+
task_categories:
|
4 |
+
- text2text-generation
|
5 |
+
language:
|
6 |
+
- zh
|
7 |
+
size_categories:
|
8 |
+
- 100M<n<1B
|
9 |
---
|
10 |
+
|
11 |
+
# Dataset for CSC
|
12 |
+
|
13 |
+
中文纠错数据集
|
14 |
+
|
15 |
+
# Dataset Description
|
16 |
+
Chinese Spelling Correction (CSC) is a task to detect and correct misspelled characters in Chinese texts.
|
17 |
+
|
18 |
+
共计 326w 条数据,以下是数据来源
|
19 |
+
|
20 |
+
|数据集|语料|链接|
|
21 |
+
|------|------|------|
|
22 |
+
|SIGHAN+Wang271K 拼写纠错数据集|SIGHAN+Wang271K(27万条)|https://huggingface.co/datasets/shibing624/CSC|
|
23 |
+
|ECSpell 拼写纠错数据集|包含法律、医疗、金融等领域|https://github.com/Aopolin-Lv/ECSpell|
|
24 |
+
|CGED 语法纠错数据集|仅包含了2016和2021年的数据集|https://github.com/wdimmy/Automatic-Corpus-Generation?spm=a2c22.12282016.0.0.5f3e7398w7SL4P|
|
25 |
+
|NLPCC 纠错数据集|包含语法纠错和拼写纠错|https://github.com/Arvid-pku/NLPCC2023_Shared_Task8;http://tcci.ccf.org.cn/conference/2023/dldoc/nacgec_training.zip;http://tcci.ccf.org.cn/conference/2018/dldoc/trainingdata02.tar.gz|
|
26 |
+
|pycorrector 语法纠错集|中文语法纠错数据集|https://github.com/shibing624/pycorrector/tree/llm/examples/data/grammar|
|
27 |
+
|
28 |
+
其余的数据集还可以看
|
29 |
+
|
30 |
+
- 中文文本纠错数据集汇总 (天池):https://tianchi.aliyun.com/dataset/138195
|
31 |
+
- NLPCC 2023中文语法纠错数据集:http://tcci.ccf.org.cn/conference/2023/taskdata.php
|
32 |
+
|
33 |
+
|
34 |
+
# Languages
|
35 |
+
The data in CSC are in Chinese.
|
36 |
+
|
37 |
+
# Dataset Structure
|
38 |
+
An example of "train" looks as follows:
|
39 |
+
```json
|
40 |
+
{"conversations":
|
41 |
+
[
|
42 |
+
{"from":"human","value":"对这个句子纠错\n\n以后,我一直以来自学汉语了。"},
|
43 |
+
{"from":"gpt","value":"从此以后,我就一直自学汉语了。"}
|
44 |
+
]
|
45 |
+
}
|
46 |
+
|
47 |
+
```
|
48 |
+
|
49 |
+
# Contributions
|
50 |
+
|
51 |
+
[Weaxs](https://github.com/Weaxs) 整理并上传
|