Update README.md
Browse files
README.md
CHANGED
|
@@ -7,4 +7,336 @@ tags:
|
|
| 7 |
pipeline_tag: token-classification
|
| 8 |
base_model:
|
| 9 |
- team-lucid/deberta-v3-small-korean
|
| 10 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
pipeline_tag: token-classification
|
| 8 |
base_model:
|
| 9 |
- team-lucid/deberta-v3-small-korean
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
## Intro
|
| 15 |
+
|
| 16 |
+
|
| 17 |
+

|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
GLiNER is a Named Entity Recognition (NER) model capable of identifying any entity type using a bidirectional transformer encoders (BERT-like). It provides a practical alternative to traditional NER models, which are limited to predefined entities, and Large Language Models (LLMs) that, despite their flexibility, are costly and large for resource-constrained scenarios.
|
| 22 |
+
|
| 23 |
+
This particular version utilize bi-encoder architecture, where textual encoder isย [team-lucid/DeBERTa v3 small](team-lucid/deberta-v3-small-korean)ย and entity label encoder is sentence transformer -ย [BAAI/bge-m3](https://huggingface.co/BAAI/bge-m3).
|
| 24 |
+
|
| 25 |
+
Such architecture brings several advantages over uni-encoder GLiNER:
|
| 26 |
+
|
| 27 |
+
- An unlimited amount of entities can be recognized at a single time;
|
| 28 |
+
- Faster inference if entity embeddings are preprocessed;
|
| 29 |
+
- Better generalization to unseen entities;
|
| 30 |
+
|
| 31 |
+
However, it has some drawbacks such as a lack of inter-label interactions that make it hard for the model to disambiguate semantically similar but contextually different entities.
|
| 32 |
+
|
| 33 |
+
|
| 34 |
+
- Paper:ย [https://arxiv.org/abs/2311.08526](https://arxiv.org/abs/2311.08526)
|
| 35 |
+
- Repository:ย [https://github.com/urchade/GLiNER](https://github.com/urchade/GLiNER)
|
| 36 |
+
- Service: https://github.com/henrikalbihn/gliner-as-a-service
|
| 37 |
+
|
| 38 |
+
---
|
| 39 |
+
## Installation & Usage
|
| 40 |
+
|
| 41 |
+
Install or update the gliner package:
|
| 42 |
+
|
| 43 |
+
```bash
|
| 44 |
+
pip install gliner>=0.2.16
|
| 45 |
+
pip install python-mecab-ko
|
| 46 |
+
```
|
| 47 |
+
|
| 48 |
+
Once you've downloaded the GLiNER library, you can import the GLiNER class. You can then load this model usingย `GLiNER.from_pretrained`ย and predict entities withย `predict_entities`.
|
| 49 |
+
|
| 50 |
+
```python
|
| 51 |
+
from gliner import GLiNER
|
| 52 |
+
|
| 53 |
+
model = GLiNER.from_pretrained("lots-o/gliner-bi-ko-small-v1")
|
| 54 |
+
|
| 55 |
+
text = """ํฌ๋ฆฌ์คํ ํผ ๋๋(Christopher Nolan) ์ ์๊ตญ์ ์ํ ๊ฐ๋
, ๊ฐ๋ณธ๊ฐ, ์ํ ํ๋ก๋์์ด๋ค. ๊ทธ์ ๋ํ์์ผ๋ก๋ 2008๋
๊ฐ๋ดํ ใ๋คํฌ ๋์ดํธใ ์๋ฆฌ์ฆ๊ฐ ์์ผ๋ฉฐ, ํนํ ใ๋คํฌ ๋์ดํธใ(2008)์ ๊ฐ๋
์ผ๋ก ๊ฐ์ฅ ์ ๋ช
ํ๋ค. ์ด ์ํ๋ ๋ฐฐํธ๋งจ ์บ๋ฆญํฐ๋ฅผ ์ค์ฌ์ผ๋ก ํ ์ํผํ์ด๋ก ์ํ๋ก, ํ์ค ๋ ์ ์ ์กฐ์ปค ์ญํ ์ด ํฐ ์ธ๊ธฐ๋ฅผ ๋์๋ค. ๋ํ, 2010๋
์ ๊ฐ๋ดํ ใ์ธ์
์
ใ(2010)์ ๋ณต์กํ ์๊ฐ๊ณผ ๊ฟ์ ๊ฐ๋
์ ๋ค๋ฃฌ SF ์ํ๋ก, ์ํ ์ ์ ๋ฐฉ์๊ณผ ์คํ ๋ฆฌ ์ ๊ฐ์์ ํ์ ์ ์ธ ์ ๊ทผ์ ์ ๋ณด์๋ค. ํฌ๋ฆฌ์คํ ํผ ๋๋์ ์๊ฐ ์ฌํ๊ณผ ๋ค์ฐจ์์ ์ด์ผ๊ธฐ๋ฅผ ํ๊ตฌํ๋ ์ํ๋ค์ ํตํด ํ๋ ์ํ๊ณ์์ ์ค์ํ ๊ฐ๋
์ผ๋ก ์๋ฆฌ๋งค๊นํ๋ค.
|
| 56 |
+
"""
|
| 57 |
+
|
| 58 |
+
labels = [
|
| 59 |
+
"์ํ/์์ค ์ํ๋ช
",
|
| 60 |
+
"์ฌ๋ ์ด๋ฆ",
|
| 61 |
+
"์บ๋ฆญํฐ ์ด๋ฆ",
|
| 62 |
+
"์ง์
๋ช
",
|
| 63 |
+
"๋ ์ง_์ฐ(๋
)",
|
| 64 |
+
"๋ ์ง_์ผ",
|
| 65 |
+
"๋ ์ง_๋ฌ(์)",
|
| 66 |
+
"๊ตญ๊ฐ"
|
| 67 |
+
]
|
| 68 |
+
|
| 69 |
+
entities = model.predict_entities(text, labels, threshold=0.2)
|
| 70 |
+
|
| 71 |
+
for entity in entities:
|
| 72 |
+
print(entity["text"], "=>", entity["label"])
|
| 73 |
+
```
|
| 74 |
+
|
| 75 |
+
```
|
| 76 |
+
ํฌ๋ฆฌ์คํ ํผ ๋๋ => ์ฌ๋ ์ด๋ฆ
|
| 77 |
+
Christopher Nolan => ์ฌ๋ ์ด๋ฆ
|
| 78 |
+
์๊ตญ => ๊ตญ๊ฐ
|
| 79 |
+
์ํ ๊ฐ๋
=> ์ง์
๋ช
|
| 80 |
+
๊ฐ๋ณธ๊ฐ => ์ง์
๋ช
|
| 81 |
+
์ํ ํ๋ก๋์ => ์ง์
๋ช
|
| 82 |
+
2008๋
=> ๋ ์ง_์ฐ(๋
)
|
| 83 |
+
๋คํฌ ๋์ดํธ => ์ํ/์์ค ์ํ๋ช
|
| 84 |
+
๋คํฌ ๋์ดํธ => ์ํ/์์ค ์ํ๋ช
|
| 85 |
+
2008 => ๋ ์ง_์ฐ(๋
)
|
| 86 |
+
๊ฐ๋
=> ์ง์
๋ช
|
| 87 |
+
๋ฐฐํธ๋งจ => ์บ๋ฆญํฐ ์ด๋ฆ
|
| 88 |
+
ํ์ค ๋ ์ => ์ฌ๋ ์ด๋ฆ
|
| 89 |
+
์กฐ์ปค => ์บ๋ฆญํฐ ์ด๋ฆ
|
| 90 |
+
2010๋
=> ๋ ์ง_์ฐ(๋
)
|
| 91 |
+
์ธ์
์
=> ์ํ/์์ค ์ํ๋ช
|
| 92 |
+
2010 => ๋ ์ง_์ฐ(๋
)
|
| 93 |
+
ํฌ๋ฆฌ์คํ ํผ ๋๋ => ์ฌ๋ ์ด๋ฆ
|
| 94 |
+
๊ฐ๋
=> ์ง์
๋ช
|
| 95 |
+
```
|
| 96 |
+
|
| 97 |
+
If you have a large amount of entities and want to pre-embed them, please, refer to the following code snippet:
|
| 98 |
+
|
| 99 |
+
```python
|
| 100 |
+
labels = ["your entities"]
|
| 101 |
+
texts = ["your texts"]
|
| 102 |
+
|
| 103 |
+
entity_embeddings = model.encode_labels(labels, batch_size = 8)
|
| 104 |
+
|
| 105 |
+
outputs = model.batch_predict_with_embeds(texts, entity_embeddings, labels)
|
| 106 |
+
```
|
| 107 |
+
|
| 108 |
+
|
| 109 |
+
---
|
| 110 |
+
|
| 111 |
+
## Dataset
|
| 112 |
+
- [๊ตญ๋ฆฝ๊ตญ์ด์ ๋ชจ๋์ ๋ง๋ญ์น](https://kli.korean.go.kr/corpus/main/requestMain.do?lang=ko)
|
| 113 |
+
- [ํ๊ตญ์ด ์ค์ฒฉ ๊ฐ์ฒด๋ช
๋ง๋ญ์น(Korean Nested Named Entity Corpus)](https://github.com/korean-named-entity/konne)
|
| 114 |
+
|
| 115 |
+
[TTA 150](https://www.korean.go.kr/front/reportData/reportDataView.do?mn_id=207&searchOrder=date&report_seq=1078&pageIndex=1)
|
| 116 |
+
|
| 117 |
+
```python
|
| 118 |
+
entity_type_mapping = {
|
| 119 |
+
"PS": {
|
| 120 |
+
"PS_NAME": "์ธ๋ฌผ_์ฌ๋",
|
| 121 |
+
"PS_CHARACTER": "์ธ๋ฌผ_๊ฐ์ ์บ๋ฆญํฐ",
|
| 122 |
+
"PS_PET": "์ธ๋ฌผ_๋ฐ๋ ค๋๋ฌผ",
|
| 123 |
+
},
|
| 124 |
+
"FD": {
|
| 125 |
+
"FD_SCIENCE": "ํ๋ฌธ ๋ถ์ผ_๊ณผํ",
|
| 126 |
+
"FD_SOCIAL_SCIENCE": "ํ๋ฌธ ๋ถ์ผ_์ฌํ๊ณผํ",
|
| 127 |
+
"FD_MEDICINE": "ํ๋ฌธ ๋ถ์ผ_์ํ",
|
| 128 |
+
"FD_ART": "ํ๋ฌธ ๋ถ์ผ_์์ ",
|
| 129 |
+
"FD_HUMANITIES": "ํ๋ฌธ ๋ถ์ผ_์ธ๋ฌธํ",
|
| 130 |
+
"FD_OTHERS": "ํ๋ฌธ ๋ถ์ผ_๊ธฐํ",
|
| 131 |
+
},
|
| 132 |
+
"TR": {
|
| 133 |
+
"TR_SCIENCE": "์ด๋ก _๊ณผํ",
|
| 134 |
+
"TR_SOCIAL_SCIENCE": "์ด๋ก _์ฌํ๊ณผํ",
|
| 135 |
+
"TR_MEDICINE": "์ด๋ก _์ํ",
|
| 136 |
+
"TR_ART": "์ด๋ก _์์ ",
|
| 137 |
+
"TR_HUMANITIES": "์ด๋ก _์ฒ ํ/์ธ์ด/์ญ์ฌ",
|
| 138 |
+
"TR_OTHERS": "์ด๋ก _๊ธฐํ",
|
| 139 |
+
},
|
| 140 |
+
"AF": {
|
| 141 |
+
"AF_BUILDING": "์ธ๊ณต๋ฌผ_๊ฑด์ถ๋ฌผ/ํ ๋ชฉ๊ฑด์ค๋ฌผ",
|
| 142 |
+
"AF_CULTURAL_ASSET": "์ธ๊ณต๋ฌผ_๋ฌธํ์ฌ",
|
| 143 |
+
"AF_ROAD": "์ธ๊ณต๋ฌผ_๋๋ก/์ฒ ๋ก",
|
| 144 |
+
"AF_TRANSPORT": "์ธ๊ณต๋ฌผ_๊ตํต์๋จ/์ด์ก์๋จ",
|
| 145 |
+
"AF_MUSICAL_INSTRUMENT": "์ธ๊ณต๋ฌผ_์
๊ธฐ",
|
| 146 |
+
"AF_WEAPON": "์ธ๊ณต๋ฌผ_๋ฌด๊ธฐ",
|
| 147 |
+
"AFA_DOCUMENT": "์ธ๊ณต๋ฌผ_๋์/์์ ์ํ๋ช
",
|
| 148 |
+
"AFA_PERFORMANCE": "์ธ๊ณต๋ฌผ_์ถค/๊ณต์ฐ/์ฐ๊ทน ์ํ๋ช
",
|
| 149 |
+
"AFA_VIDEO": "์ธ๊ณต๋ฌผ_์ํ/TV ํ๋ก๊ทธ๋จ",
|
| 150 |
+
"AFA_ART_CRAFT": "์ธ๊ณต๋ฌผ_๋ฏธ์ /์กฐํ ์ํ๋ช
",
|
| 151 |
+
"AFA_MUSIC": "์ธ๊ณต๋ฌผ_์์
์ํ๋ช
",
|
| 152 |
+
"AFW_SERVICE_PRODUCTS": "์ธ๊ณต๋ฌผ_์๋น์ค ์ํ",
|
| 153 |
+
"AFW_OTHER_PRODUCTS": "์ธ๊ณต๋ฌผ_๊ธฐํ ์ํ",
|
| 154 |
+
},
|
| 155 |
+
"OG": {
|
| 156 |
+
"OGG_ECONOMY": "๊ธฐ๊ด_๊ฒฝ์ ",
|
| 157 |
+
"OGG_EDUCATION": "๊ธฐ๊ด_๊ต์ก",
|
| 158 |
+
"OGG_MILITARY": "๊ธฐ๊ด_๊ตฐ์ฌ",
|
| 159 |
+
"OGG_MEDIA": "๊ธฐ๊ด_๋ฏธ๋์ด",
|
| 160 |
+
"OGG_SPORTS": "๊ธฐ๊ด_์คํฌ์ธ ",
|
| 161 |
+
"OGG_ART": "๊ธฐ๊ด_์์ ",
|
| 162 |
+
"OGG_MEDICINE": "๊ธฐ๊ด_์๋ฃ",
|
| 163 |
+
"OGG_RELIGION": "๊ธฐ๊ด_์ข
๊ต",
|
| 164 |
+
"OGG_SCIENCE": "๊ธฐ๊ด_๊ณผํ",
|
| 165 |
+
"OGG_LIBRARY": "๊ธฐ๊ด_๋์๊ด",
|
| 166 |
+
"OGG_LAW": "๊ธฐ๊ด_๋ฒ๋ฅ ",
|
| 167 |
+
"OGG_POLITICS": "๊ธฐ๊ด_์ ๋ถ/๊ณต๊ณต",
|
| 168 |
+
"OGG_FOOD": "๊ธฐ๊ด_์์ ์
์ฒด",
|
| 169 |
+
"OGG_HOTEL": "๊ธฐ๊ด_์๋ฐ ์
์ฒด",
|
| 170 |
+
"OGG_OTHERS": "๊ธฐ๊ด_๊ธฐํ",
|
| 171 |
+
},
|
| 172 |
+
"LC": {
|
| 173 |
+
"LCP_COUNTRY": "์ฅ์_๊ตญ๊ฐ",
|
| 174 |
+
"LCP_PROVINCE": "์ฅ์_๋/์ฃผ ์ง์ญ",
|
| 175 |
+
"LCP_COUNTY": "์ฅ์_์ธ๋ถ ํ์ ๊ตฌ์ญ",
|
| 176 |
+
"LCP_CITY": "์ฅ์_๋์",
|
| 177 |
+
"LCP_CAPITALCITY": "์ฅ์_์๋",
|
| 178 |
+
"LCG_RIVER": "์ฅ์_๊ฐ/ํธ์",
|
| 179 |
+
"LCG_OCEAN": "์ฅ์_๋ฐ๋ค",
|
| 180 |
+
"LCG_BAY": "์ฅ์_๋ฐ๋/๋ง",
|
| 181 |
+
"LCG_MOUNTAIN": "์ฅ์_์ฐ/์ฐ๋งฅ",
|
| 182 |
+
"LCG_ISLAND": "์ฅ์_์ฌ",
|
| 183 |
+
"LCG_CONTINENT": "์ฅ์_๋๋ฅ",
|
| 184 |
+
"LC_SPACE": "์ฅ์_์ฒ์ฒด",
|
| 185 |
+
"LC_OTHERS": "์ฅ์_๊ธฐํ",
|
| 186 |
+
},
|
| 187 |
+
"CV": {
|
| 188 |
+
"CV_CULTURE": "๋ฌธ๋ช
_๋ฌธ๋ช
/๋ฌธํ",
|
| 189 |
+
"CV_TRIBE": "๋ฌธ๋ช
_๋ฏผ์กฑ/์ข
์กฑ",
|
| 190 |
+
"CV_LANGUAGE": "๋ฌธ๋ช
_์ธ์ด",
|
| 191 |
+
"CV_POLICY": "๋ฌธ๋ช
_์ ๋/์ ์ฑ
",
|
| 192 |
+
"CV_LAW": "๋ฌธ๋ช
_๋ฒ/๋ฒ๋ฅ ",
|
| 193 |
+
"CV_CURRENCY": "๋ฌธ๋ช
_ํตํ",
|
| 194 |
+
"CV_TAX": "๋ฌธ๋ช
_์กฐ์ธ",
|
| 195 |
+
"CV_FUNDS": "๋ฌธ๋ช
_์ฐ๊ธ/๊ธฐ๊ธ",
|
| 196 |
+
"CV_ART": "๋ฌธ๋ช
_์์ ",
|
| 197 |
+
"CV_SPORTS": "๋ฌธ๋ช
_์คํฌ์ธ ",
|
| 198 |
+
"CV_SPORTS_POSITION": "๋ฌธ๋ช
_์คํฌ์ธ ํฌ์ง์
",
|
| 199 |
+
"CV_SPORTS_INST": "๋ฌธ๋ช
_์คํฌ์ธ ์ฉํ/๋๊ตฌ",
|
| 200 |
+
"CV_PRIZE": "๋ฌธ๋ช
_์/ํ์ฅ",
|
| 201 |
+
"CV_RELATION": "๋ฌธ๋ช
_๊ฐ์กฑ/์น์กฑ ๊ด๊ณ",
|
| 202 |
+
"CV_OCCUPATION": "๋ฌธ๋ช
_์ง์
",
|
| 203 |
+
"CV_POSITION": "๋ฌธ๋ช
_์ง์/์ง์ฑ
",
|
| 204 |
+
"CV_FOOD": "๋ฌธ๋ช
_์์",
|
| 205 |
+
"CV_DRINK": "๋ฌธ๋ช
_์๋ฃ/์ ",
|
| 206 |
+
"CV_FOOD_STYLE": "๋ฌธ๋ช
_์์ ์ ํ",
|
| 207 |
+
"CV_CLOTHING": "๋ฌธ๋ช
_์๋ณต/์ฌ์ ",
|
| 208 |
+
"CV_BUILDING_TYPE": "๋ฌธ๋ช
_๊ฑด์ถ ์์",
|
| 209 |
+
},
|
| 210 |
+
"DT": {
|
| 211 |
+
"DT_DURATION": "๋ ์ง_๊ธฐ๊ฐ",
|
| 212 |
+
"DT_DAY": "๋ ์ง_์ผ",
|
| 213 |
+
"DT_WEEK": "๋ ์ง_์ฃผ(์ฃผ์ฐจ)",
|
| 214 |
+
"DT_MONTH": "๋ ์ง_๋ฌ(์)",
|
| 215 |
+
"DT_YEAR": "๋ ์ง_์ฐ(๋
)",
|
| 216 |
+
"DT_SEASON": "๋ ์ง_๊ณ์ ",
|
| 217 |
+
"DT_GEOAGE": "๋ ์ง_์ง์ง์๋",
|
| 218 |
+
"DT_DYNASTY": "๋ ์ง_์์กฐ์๋",
|
| 219 |
+
"DT_OTHERS": "๋ ์ง_๊ธฐํ",
|
| 220 |
+
},
|
| 221 |
+
"TI": {
|
| 222 |
+
"TI_DURATION": "์๊ฐ_๊ธฐ๊ฐ",
|
| 223 |
+
"TI_HOUR": "์๊ฐ_์๊ฐ(์)",
|
| 224 |
+
"TI_MINUTE": "์๊ฐ_๋ถ",
|
| 225 |
+
"TI_SECOND": "์๊ฐ_์ด",
|
| 226 |
+
"TI_OTHERS": "์๊ฐ_๊ธฐํ",
|
| 227 |
+
},
|
| 228 |
+
"QT": {
|
| 229 |
+
"QT_AGE": "์๋_๋์ด",
|
| 230 |
+
"QT_SIZE": "์๋_๋์ด/๋ฉด์ ",
|
| 231 |
+
"QT_LENGTH": "์๋_๊ธธ์ด/๊ฑฐ๋ฆฌ",
|
| 232 |
+
"QT_COUNT": "์๋_์๋/๋น๋",
|
| 233 |
+
"QT_MAN_COUNT": "์๋_์ธ์์",
|
| 234 |
+
"QT_WEIGHT": "์๋_๋ฌด๊ฒ",
|
| 235 |
+
"QT_PERCENTAGE": "์๋_๋ฐฑ๋ถ์จ",
|
| 236 |
+
"QT_SPEED": "์๋_์๋",
|
| 237 |
+
"QT_TEMPERATURE": "์๋_์จ๋",
|
| 238 |
+
"QT_VOLUME": "์๋_๋ถํผ",
|
| 239 |
+
"QT_ORDER": "์๋_์์",
|
| 240 |
+
"QT_PRICE": "์๋_๊ธ์ก",
|
| 241 |
+
"QT_PHONE": "์๋_์ ํ๋ฒํธ",
|
| 242 |
+
"QT_SPORTS": "์๋_์คํฌ์ธ ์๋",
|
| 243 |
+
"QT_CHANNEL": "์๋_์ฑ๋ ๋ฒํธ",
|
| 244 |
+
"QT_ALBUM": "์๋_์จ๋ฒ ์๋",
|
| 245 |
+
"QT_ADDRESS": "์๋_์ฃผ์ ๊ด๋ จ ์ซ์",
|
| 246 |
+
"QT_OTHERS": "์๋_๊ธฐํ ์๋",
|
| 247 |
+
},
|
| 248 |
+
"EV": {
|
| 249 |
+
"EV_ACTIVITY": "์ฌ๊ฑด_์ฌํ์ด๋/์ ์ธ",
|
| 250 |
+
"EV_WAR_REVOLUTION": "์ฌ๊ฑด_์ ์/ํ๋ช
",
|
| 251 |
+
"EV_SPORTS": "์ฌ๊ฑด_์คํฌ์ธ ํ์ฌ",
|
| 252 |
+
"EV_FESTIVAL": "์ฌ๊ฑด_์ถ์ /์ํ์ ",
|
| 253 |
+
"EV_OTHERS": "์ฌ๊ฑด_๊ธฐํ",
|
| 254 |
+
},
|
| 255 |
+
"AM": {
|
| 256 |
+
"AM_INSECT": "๋๋ฌผ_๊ณค์ถฉ",
|
| 257 |
+
"AM_BIRD": "๋๋ฌผ_์กฐ๋ฅ",
|
| 258 |
+
"AM_FISH": "๋๋ฌผ_์ด๋ฅ",
|
| 259 |
+
"AM_MAMMALIA": "๋๋ฌผ_ํฌ์ ๋ฅ",
|
| 260 |
+
"AM_AMPHIBIA": "๋๋ฌผ_์์๋ฅ",
|
| 261 |
+
"AM_REPTILIA": "๋๋ฌผ_ํ์ถฉ๋ฅ",
|
| 262 |
+
"AM_TYPE": "๋๋ฌผ_๋ถ๋ฅ๋ช
",
|
| 263 |
+
"AM_PART": "๋๋ฌผ_๋ถ์๋ช
",
|
| 264 |
+
"AM_OTHERS": "๋๋ฌผ_๊ธฐํ",
|
| 265 |
+
},
|
| 266 |
+
"PT": {
|
| 267 |
+
"PT_FRUIT": "์๋ฌผ_๊ณผ์ผ/์ด๋งค",
|
| 268 |
+
"PT_FLOWER": "์๋ฌผ_๊ฝ",
|
| 269 |
+
"PT_TREE": "์๋ฌผ_๋๋ฌด",
|
| 270 |
+
"PT_GRASS": "์๋ฌผ_ํ",
|
| 271 |
+
"PT_TYPE": "์๋ฌผ_๋ถ๋ฅ๋ช
",
|
| 272 |
+
"PT_PART": "์๋ฌผ_๋ถ์๋ช
",
|
| 273 |
+
"PT_OTHERS": "์๋ฌผ_๊ธฐํ",
|
| 274 |
+
},
|
| 275 |
+
"MT": {
|
| 276 |
+
"MT_ELEMENT": "๋ฌผ์ง_์์",
|
| 277 |
+
"MT_METAL": "๋ฌผ์ง_๊ธ์",
|
| 278 |
+
"MT_ROCK": "๋ฌผ์ง_์์",
|
| 279 |
+
"MT_CHEMICAL": "๋ฌผ์ง_ํํ",
|
| 280 |
+
},
|
| 281 |
+
"TM": {
|
| 282 |
+
"TM_COLOR": "์ฉ์ด_์๊น",
|
| 283 |
+
"TM_DIRECTION": "์ฉ์ด_๋ฐฉํฅ",
|
| 284 |
+
"TM_CLIMATE": "์ฉ์ด_๊ธฐํ ์ง์ญ",
|
| 285 |
+
"TM_SHAPE": "์ฉ์ด_๋ชจ์/ํํ",
|
| 286 |
+
"TM_CELL_TISSUE_ORGAN": "์ฉ์ด_์ธํฌ/์กฐ์ง/๊ธฐ๊ด",
|
| 287 |
+
"TMM_DISEASE": "์ฉ์ด_์ฆ์/์ง๋ณ",
|
| 288 |
+
"TMM_DRUG": "์ฉ์ด_์ฝํ",
|
| 289 |
+
"TMI_HW": "์ฉ์ด_IT ํ๋์จ์ด",
|
| 290 |
+
"TMI_SW": "์ฉ์ด_IT ์ํํธ์จ์ด",
|
| 291 |
+
"TMI_SITE": "์ฉ์ด_URL ์ฃผ์",
|
| 292 |
+
"TMI_EMAIL": "์ฉ์ด_์ด๋ฉ์ผ ์ฃผ์",
|
| 293 |
+
"TMI_MODEL": "์ฉ์ด_์ ํ ๋ชจ๋ธ๋ช
",
|
| 294 |
+
"TMI_SERVICE": "์ฉ์ด_IT ์๋น์ค",
|
| 295 |
+
"TMI_PROJECT": "์ฉ์ด_ํ๋ก์ ํธ",
|
| 296 |
+
"TMIG_GENRE": "์ฉ์ด_๊ฒ์ ์ฅ๋ฅด",
|
| 297 |
+
"TM_SPORTS": "์ฉ์ด_์คํฌ์ธ ",
|
| 298 |
+
},
|
| 299 |
+
}
|
| 300 |
+
```
|
| 301 |
+
## Evaluation
|
| 302 |
+
|
| 303 |
+
|
| 304 |
+
Evaluate with theย [konne dev set](https://github.com/korean-named-entity/konne) :
|
| 305 |
+
The evaluation results presented in the table below, except for the values I provided, were derived from the following source:ย [taeminlee/gliner_ko.](https://huggingface.co/taeminlee/gliner_ko)
|
| 306 |
+
|
| 307 |
+
| Model | Precision(P) | Recall(R) | F1 |
|
| 308 |
+
| :----------------------------: | :----------: | :-------: | :--------: |
|
| 309 |
+
| gliner-bi-ko-small-v1 (t=0.5) | 81.53% | 74.16% | 77.67% |
|
| 310 |
+
| gliner-bi-ko-xlarge-v1 (t=0.5) | **84.73%** | 77.71% | **81.07%** |
|
| 311 |
+
| Gliner-ko (t=0.5) | 72.51% | 79.82% | 75.99% |
|
| 312 |
+
| Gliner Large-v2 (t=0.5) | 34.33% | 19.50% | 24.87% |
|
| 313 |
+
| Gliner Multi (t=0.5) | 40.94% | 34.18% | 37.26% |
|
| 314 |
+
| Pororo | 70.25% | 57.94% | 63.50% |
|
| 315 |
+
|
| 316 |
+
|
| 317 |
+
|
| 318 |
+
---
|
| 319 |
+
|
| 320 |
+
## Citation
|
| 321 |
+
```bibtex
|
| 322 |
+
@misc{gliner_bi_ko_small_v1,
|
| 323 |
+
title={gliner-bi-ko-small-v1},
|
| 324 |
+
author={Gihwan Kim},
|
| 325 |
+
year={2025},
|
| 326 |
+
url={https://huggingface.co/lots-o/gliner-bi-ko-small-v1}
|
| 327 |
+
publisher={Hugging Face}
|
| 328 |
+
}
|
| 329 |
+
|
| 330 |
+
```
|
| 331 |
+
|
| 332 |
+
```bibtex
|
| 333 |
+
@misc{zaratiana2023gliner,
|
| 334 |
+
title={GLiNER: Generalist Model for Named Entity Recognition using Bidirectional Transformer},
|
| 335 |
+
author={Urchade Zaratiana and Nadi Tomeh and Pierre Holat and Thierry Charnois},
|
| 336 |
+
year={2023},
|
| 337 |
+
eprint={2311.08526},
|
| 338 |
+
archivePrefix={arXiv},
|
| 339 |
+
primaryClass={cs.CL}
|
| 340 |
+
}
|
| 341 |
+
```
|
| 342 |
+
|