|
--- |
|
annotations_creators: |
|
- expert-generated |
|
- machine-generated |
|
language_creators: |
|
- found |
|
- expert-generated |
|
languages: |
|
- th |
|
licenses: |
|
- cc-by-3-0 |
|
multilinguality: |
|
- monolingual |
|
size_categories: |
|
- 1k<n<10k |
|
source_datasets: |
|
- extended|other-tirasaroj-aroonmanakun |
|
task_categories: |
|
- structure-prediction |
|
task_ids: |
|
- named-entity-recognition |
|
- parsing |
|
--- |
|
|
|
# Dataset Card for `thainer` |
|
|
|
## Table of Contents |
|
- [Dataset Description](#dataset-description) |
|
- [Dataset Summary](#dataset-summary) |
|
- [Supported Tasks](#supported-tasks-and-leaderboards) |
|
- [Languages](#languages) |
|
- [Dataset Structure](#dataset-structure) |
|
- [Data Instances](#data-instances) |
|
- [Data Fields](#data-instances) |
|
- [Data Splits](#data-instances) |
|
- [Dataset Creation](#dataset-creation) |
|
- [Curation Rationale](#curation-rationale) |
|
- [Source Data](#source-data) |
|
- [Annotations](#annotations) |
|
- [Personal and Sensitive Information](#personal-and-sensitive-information) |
|
- [Considerations for Using the Data](#considerations-for-using-the-data) |
|
- [Social Impact of Dataset](#social-impact-of-dataset) |
|
- [Discussion of Biases](#discussion-of-biases) |
|
- [Other Known Limitations](#other-known-limitations) |
|
- [Additional Information](#additional-information) |
|
- [Dataset Curators](#dataset-curators) |
|
- [Licensing Information](#licensing-information) |
|
- [Citation Information](#citation-information) |
|
|
|
## Dataset Description |
|
|
|
- **Homepage:** https://github.com/wannaphong/thai-ner |
|
- **Repository:** https://github.com/wannaphong/thai-ner |
|
- **Paper:** |
|
- **Leaderboard:** |
|
- **Point of Contact:** https://github.com/wannaphong/ |
|
|
|
### Dataset Summary |
|
|
|
ThaiNER (v1.3) is a 6,456-sentence named entity recognition dataset created from expanding the 2,258-sentence [unnamed dataset](http://pioneer.chula.ac.th/~awirote/Data-Nutcha.zip) by [Tirasaroj and Aroonmanakun (2012)](http://pioneer.chula.ac.th/~awirote/publications/). It is used to train NER taggers in [PyThaiNLP](https://github.com/PyThaiNLP/pythainlp). The NER tags are annotated by [Tirasaroj and Aroonmanakun (2012)]((http://pioneer.chula.ac.th/~awirote/publications/)) for 2,258 sentences and the rest by [@wannaphong](https://github.com/wannaphong/). The POS tags are done by [PyThaiNLP](https://github.com/PyThaiNLP/pythainlp)'s `perceptron` engine trained on `orchid_ud`. [@wannaphong](https://github.com/wannaphong/) is now the only maintainer of this dataset. |
|
|
|
### Supported Tasks and Leaderboards |
|
|
|
- named entity recognition |
|
- pos tagging |
|
|
|
### Languages |
|
|
|
Thai |
|
|
|
## Dataset Structure |
|
|
|
### Data Instances |
|
|
|
``` |
|
{'id': 100, 'ner_tags': [27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27], 'pos_tags': [6, 12, 13, 1, 6, 5, 11, 7, 11, 6, 5, 13, 6, 6, 6, 11, 6, 6, 11, 6, 6, 11, 6, 6, 13, 6, 11, 11, 6, 11, 6, 11, 6, 11, 6, 11, 11, 6, 6, 11, 12, 6, 13, 5, 11, 7, 11, 6, 3, 11, 12, 3, 13, 6, 1, 6, 12, 13, 1, 6, 6, 5, 11, 3, 11, 5, 4, 6, 13, 6, 13, 6, 10, 3, 13, 13, 12, 13, 12, 0, 1, 10, 11, 6, 6, 11, 6, 11, 6, 12, 13, 5, 12, 3, 13, 13, 1, 6, 1, 6, 13], 'tokens': ['เชื้อโรค', 'ที่', 'ปรากฏ', 'ใน', 'สัตว์', 'ทั้ง', ' ', '4', ' ', 'ชนิด', 'นี้', 'เป็น', 'เชื้อ', 'โรคไข้หวัด', 'นก', ' ', 'เอช', 'พี', ' ', 'เอ', 'เวียน', ' ', 'อิน', 'ฟลู', 'เอน', 'ซา', ' ', '(', 'Hight', ' ', 'Polygenic', ' ', 'Avain', ' ', 'Influenza', ')', ' ', 'ชนิด', 'รุนแรง', ' ', 'ซึ่ง', 'การ', 'ตั้งชื่อ', 'ทั้ง', ' ', '4', ' ', 'ขึ้น', 'มา', ' ', 'เพื่อที่จะ', 'สามารถ', 'ระบุ', 'เชื้อ', 'ของ', 'ไวรัส', 'ที่', 'ทำอันตราย', 'ตาม', 'สิ่งมีชีวิต', 'ประเภท', 'ต่างๆ', ' ', 'ได้', ' ', 'อีก', 'ทั้ง', 'การ', 'ระบุ', 'สถานที่', 'คือ', 'ประเทศ', 'ไทย', 'จะ', 'ทำให้', 'รู้', 'ว่า', 'พบ', 'ที่', 'แรก', 'ใน', 'ไทย', ' ', 'ส่วน', 'วัน', ' ', 'เดือน', ' ', 'ปี', 'ที่', 'พบ', 'นั้น', 'ก็', 'จะ', 'ทำให้', 'ทราบ', 'ถึง', 'ครั้งแรก', 'ของ', 'การ', 'ค้นพบ']} |
|
{'id': 107, 'ner_tags': [27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27], 'pos_tags': [0, 1, 6, 5, 11, 12, 3, 3, 13, 6, 13, 12, 0, 2, 12, 11, 6, 5, 13, 6, 5, 1, 6, 6, 1, 10, 11, 4, 13, 6, 11, 12, 6, 6, 10, 11, 13, 6, 1, 6, 4, 6, 1, 6, 6, 11, 4, 6, 1, 5, 6, 12, 2, 13, 6, 6, 5, 1, 11, 12, 13, 1, 6, 6, 11, 13, 11, 6, 6, 6, 11, 11, 6, 11, 11, 4, 10, 11, 11, 6, 11], 'tokens': ['ล่าสุด', 'ใน', 'เรื่อง', 'นี้', ' ', 'ทั้งนี้', 'คง', 'ต้อง', 'มี', 'การ', 'ตรวจสอบ', 'ให้', 'ชัดเจน', 'อีกครั้ง', 'ว่า', ' ', 'ไวรัส', 'นี้', 'เป็น', 'ชนิด', 'เดียว', 'กับ', 'ไข้หวัด', 'นก', 'ใน', 'ไทย', ' ', 'หรือ', 'เป็น', 'การกลายพันธุ์', ' ', 'โดยที่', 'คณะ', 'สัตวแพทย์', 'มหาวิทยาลัยเกษตรศาสตร์', ' ', 'จัด', 'ระดมสมอง', 'จาก', 'คณบดี', 'และ', 'ผู้เชี่ยวชาญ', 'จาก', 'คณะ', 'สัตวแพทย์', ' ', 'และ', 'ปศุสัตว์', 'ของ', 'หลาย', 'มหาวิทยาลัย', 'เพื่อ', 'ร่วมกัน', 'หา', 'ข้อมูล', 'เรื่อง', 'นี้', 'ด้วย', ' ', 'โดย', 'ประสาน', 'กับ', 'เจ้าหน้าที่', 'ระหว่างประเทศ', ' ', 'คือ', ' ', 'องค์การ', 'สุขภาพ', 'สัตว์โลก', ' ', '(', 'OIE', ')', ' ', 'และ', 'องค์การอนามัยโลก', ' ', '(', 'WHO', ')']} |
|
``` |
|
|
|
### Data Fields |
|
|
|
- `id`: sentence id |
|
- `tokens`: word tokens by [PyThaiNLP](https://github.com/PyThaiNLP/pythainlp)'s dictionary-based tokenizer `newmm` |
|
- `pos_tags`: POS tags tagged by [PyThaiNLP](https://github.com/PyThaiNLP/pythainlp)'s `perceptron` engine trained on `orchid_ud` |
|
- `ner_tags`: NER tags tagged by humans |
|
|
|
### Data Splits |
|
|
|
No explicit split is given |
|
|
|
## Dataset Creation |
|
|
|
### Curation Rationale |
|
|
|
ThaiNER (v1.3) is a 6,456-sentence named entity recognition dataset created from expanding the 2,258-sentence [unnamed dataset](http://pioneer.chula.ac.th/~awirote/Data-Nutcha.zip) by [Tirasaroj and Aroonmanakun (2012)](http://pioneer.chula.ac.th/~awirote/publications/). It is used to train NER taggers in [PyThaiNLP](https://github.com/PyThaiNLP/pythainlp). |
|
|
|
### Source Data |
|
|
|
#### Initial Data Collection and Normalization |
|
|
|
The earlier part of the dataset is all news articles, whereas the part added by [@wannaphong](https://github.com/wannaphong/) includes news articles, public announcements and [@wannaphong](https://github.com/wannaphong/)'s own chat messages with personal and sensitive information removed. |
|
|
|
#### Who are the source language producers? |
|
|
|
News articles and public announcements are created by their respective authors. Chat messages are created by [@wannaphong](https://github.com/wannaphong/). |
|
|
|
### Annotations |
|
|
|
#### Annotation process |
|
|
|
[More Information Needed] |
|
|
|
#### Who are the annotators? |
|
|
|
[Tirasaroj and Aroonmanakun (2012)](http://pioneer.chula.ac.th/~awirote/publications/) for the earlier 2,258 sentences and [@wannaphong](https://github.com/wannaphong/) for the rest |
|
|
|
### Personal and Sensitive Information |
|
|
|
News articles and public announcements are not expected to include personal and sensitive information. [@wannaphong](https://github.com/wannaphong/) has removed such information from his own chat messages. |
|
|
|
## Considerations for Using the Data |
|
|
|
### Social Impact of Dataset |
|
|
|
- named entity recognition in Thai |
|
|
|
### Discussion of Biases |
|
|
|
Since almost all of collection and annotation is done by [@wannaphong](https://github.com/wannaphong/), his biases are expected to be reflected in the dataset. |
|
|
|
### Other Known Limitations |
|
|
|
[More Information Needed] |
|
|
|
## Additional Information |
|
|
|
### Dataset Curators |
|
|
|
[Tirasaroj and Aroonmanakun (2012)](http://pioneer.chula.ac.th/~awirote/publications/) for the earlier 2,258 sentences and [@wannaphong](https://github.com/wannaphong/) for the rest |
|
|
|
### Licensing Information |
|
|
|
CC-BY 3.0 |
|
|
|
### Citation Information |
|
|
|
``` |
|
@misc{Wannaphong Phatthiyaphaibun_2019, |
|
title={wannaphongcom/thai-ner: ThaiNER 1.3}, |
|
url={https://zenodo.org/record/3550546}, |
|
DOI={10.5281/ZENODO.3550546}, |
|
abstractNote={Thai Named Entity Recognition}, |
|
publisher={Zenodo}, |
|
author={Wannaphong Phatthiyaphaibun}, |
|
year={2019}, |
|
month={Nov} |
|
} |
|
``` |
|
|
|
Work extended from: |
|
[Tirasaroj, N. and Aroonmanakun, W. 2012. Thai NER using CRF model based on surface features. In Proceedings of SNLP-AOS 2011, 9-10 February, 2012, Bangkok, pages 176-180.](http://pioneer.chula.ac.th/~awirote/publications/) |
|
|
