Update README.md
Browse files
README.md
CHANGED
|
@@ -5,4 +5,109 @@ language:
|
|
| 5 |
- en
|
| 6 |
base_model:
|
| 7 |
- klue/bert-base
|
| 8 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
- en
|
| 6 |
base_model:
|
| 7 |
- klue/bert-base
|
| 8 |
+
---
|
| 9 |
+
# LQ-KBERT-Base: Crypto Market Korean Sentiment & Action Signal Classifier
|
| 10 |
+
|
| 11 |
+
[LangQuant](https://langquant.com)์์ ๊ณต๊ฐํ **ํ๊ตญ์ด ๊ธ์ต ์ปค๋ฎค๋ํฐ/๋ด์ค ํฌ์์ฌ๋ฆฌ ๋ถ๋ฅ ๋ชจ๋ธ**์
๋๋ค.
|
| 12 |
+
`klue/bert-base`๋ฅผ ๋ฐฑ๋ณธ์ผ๋ก ํ๊ณ , ๊ฐ์์์ฐ ๊ด๋ จ ํ๊ตญ์ด ๋ฐ์ดํฐ์
**10๋ง ๊ฑด ์ด์**์ ์ ์ฒ๋ฆฌํ์ฌ ํ์ธํ๋ํ์ต๋๋ค.
|
| 13 |
+
๋ชจ๋ธ์ ๋ฌธ์ฅ ๋จ์ ์
๋ ฅ(`โค200์`)์ ๋ํด **ํฌ์ ์ฌ๋ฆฌยทํ๋ยท๊ฐ์ ยทํ์ ๋ยท๊ด๋ จ์ฑยท์ ํด์ฑ**์ ๋์์ ์์ธกํฉ๋๋ค.
|
| 14 |
+
|
| 15 |
+
- [Github](https://github.com/LangQuant/LQ-KBERT-Base)
|
| 16 |
+
---
|
| 17 |
+
### ๋ชจ๋ธ์ ๋ค์ ํญ๋ชฉ์ ์์ธกํฉ๋๋ค.
|
| 18 |
+
|
| 19 |
+
```json
|
| 20 |
+
{
|
| 21 |
+
"sentiment_strength": "strong_pos | weak_pos | neutral | weak_neg | strong_neg",
|
| 22 |
+
"action_signal": "buy | hold | sell | avoid | info_only | ask_info",
|
| 23 |
+
"emotions": ["greed","fear","confidence","doubt","anger","hope","sarcasm"],
|
| 24 |
+
"certainty": 0.0 ~ 1.0,
|
| 25 |
+
"relevance": 0.0 ~ 1.0,
|
| 26 |
+
"reasons": "๋ผ๋ฒจ ๊ทผ๊ฑฐ๋ฅผ ์์ฝํ ํ๊ตญ์ด 1~2๋ฌธ์ฅ",
|
| 27 |
+
"toxicity": 0.0 ~ 1.0
|
| 28 |
+
}
|
| 29 |
+
```
|
| 30 |
+
---
|
| 31 |
+
## Labeling Guidelines
|
| 32 |
+
|
| 33 |
+
### Sentiment Strength
|
| 34 |
+
- **strong_pos**: ๊ธ๋ฑ ํ์ , `"๊ฐ์ฆ์"`, `"๋ฌด์กฐ๊ฑด ๊ฐ๋ค"`.
|
| 35 |
+
- **weak_pos**: ์กฐ์ฌ์ค๋ฌ์ด ๋๊ด, `"๋ฐ๋ฑ ๊ฐ๋ฅ"`, `"๊ด์ฐฎ์ ๋ฏ"`.
|
| 36 |
+
- **neutral**: ๋จ์ ์ ๋ณด/๊ณต์ง/์ก๋ด.
|
| 37 |
+
- **weak_neg**: ์๊ณกํ ๋ถ์ , `"์กฐ์ ์ฌ ๋ฏ"`, `"๊ด๋ง"`.
|
| 38 |
+
- **strong_neg**: ํญ๋ฝยทํจ๋, `"๋๋ฝ"`, `"๋งํจ"`, `"ํดํน/์ ์ฌ"`.
|
| 39 |
+
|
| 40 |
+
### Action Signal
|
| 41 |
+
- **buy**: ๋งค์/์ง์
์ง์, `"์ง๊ธ ์ฐ๋ค"`, `"๋กฑ"`.
|
| 42 |
+
- **hold**: ๋ณด์ ์ ์ง/๊ด๋ง, `"์กด๋ฒ"`, `"์ ์ง"`.
|
| 43 |
+
- **sell**: ๋งค๋/์ฒญ์ฐ, `"์ต์ "`, `"์์ "`, `"์ ๋ฆฌ"`.
|
| 44 |
+
- **avoid**: ํํผ/์ํ ๊ฒฝ๊ณ , `"๊ฐ์ง๋ง"`, `"์ค์บ "`, `"์ํ"`.
|
| 45 |
+
- **info_only**: ๋จ์ ์ ๋ณด ์ ๋ฌ (๋ด์ค/๊ณต์ง).
|
| 46 |
+
- **ask_info**: ์ง๋ฌธ/ํ์, `"๋ค์ด๊ฐ๋ ๋ผ?"`, `"์ ๋จ์ด์ ธ?"`.
|
| 47 |
+
|
| 48 |
+
### Emotions (๋ค์ค ์ ํ)
|
| 49 |
+
- **greed** ํ์
|
| 50 |
+
- **fear** ๋๋ ค์
|
| 51 |
+
- **confidence** ํ์
|
| 52 |
+
- **doubt** ์์ฌ
|
| 53 |
+
- **anger** ๋ถ๋
ธ
|
| 54 |
+
- **hope** ํฌ๋ง
|
| 55 |
+
- **sarcasm** ํ์
|
| 56 |
+
|
| 57 |
+
### Certainty
|
| 58 |
+
- **0.2~0.4**: ์ง๋ฌธยทํ์ยท๋ฐ (๋ฎ์)
|
| 59 |
+
- **0.4~0.6**: ์๊ณกํ ์๊ฒฌ (์ค๊ฐ)
|
| 60 |
+
- **0.6~0.8**: ์์นยท๊ทผ๊ฑฐยท๊ณต์์ฑ (๋์)
|
| 61 |
+
- **0.8~1.0**: ๊ฐํ ๋จ์ ยท์ง์ (๋งค์ฐ ๋์)
|
| 62 |
+
|
| 63 |
+
### Relevance
|
| 64 |
+
- **0.7~1.0**: ์ง์ ์ ์ธ ํฌ์/์์ฅ ๊ด๋ จ
|
| 65 |
+
- **0.4~0.7**: ๊ฐ์ ๊ด๋ จ (์
๊ณ/์ธ๋ฌผ/๊ธฐ์ )
|
| 66 |
+
- **0.0~0.3**: ๋ฌด๊ด/์ก๋ด/๋ฐ
|
| 67 |
+
|
| 68 |
+
### Toxicity
|
| 69 |
+
- ์์คยท๋ชจ์ยท๋นํ ๊ฐ๋์ ๋ฐ๋ผ **0~1**.
|
| 70 |
+
- ํฌ์ ์๋ฏธ์๋ ๋ณ๋๋ก ๋
๋ฆฝ์ ์ผ๋ก ํ๊ฐ.
|
| 71 |
+
|
| 72 |
+
---
|
| 73 |
+
|
| 74 |
+
## Sentiment Strength vs Action Signal
|
| 75 |
+
|
| 76 |
+
- **Sentiment Strength**
|
| 77 |
+
- ํฌ์ ์ฌ๋ฆฌ์ ๊ฐ๋ (๊ธ์ โ ๋ถ์ ).
|
| 78 |
+
- ๊ฐ๊ฒฉ ์ ๋ง์ ํค์ ์ง์ค.
|
| 79 |
+
|
| 80 |
+
- **Action Signal**
|
| 81 |
+
- ์ค์ ํฌ์ ํ๋ ์๋/์ง์.
|
| 82 |
+
- ๋งค์/๋งค๋/๋ณด์ /ํํผ/์ง๋ฌธ/์ ๋ณด.
|
| 83 |
+
|
| 84 |
+
|
| 85 |
+
---
|
| 86 |
+
|
| 87 |
+
### ์์
|
| 88 |
+
|
| 89 |
+
| ๋ฌธ์ฅ | sentiment_strength | action_signal | ํด์ |
|
| 90 |
+
|------|--------------------|---------------|------|
|
| 91 |
+
| "๊ฐ๋ก์์ด์ฌ " | strong_pos | buy | ๊ฐํ ์์น ํ์ + ์ฆ์ ๋งค์ ์๋ |
|
| 92 |
+
| "์ฌ๊ธฐ์ ๊ด๋ง์ด ๋ง๋ค" | weak_neg | hold | ๋ถ์ ์ ์ด์ง๋ง ๋ณด์ ์ ์ง ์ ํ |
|
| 93 |
+
| "๋ค์ด๊ฐ๋ ๋ ๊น?" | weak_pos | ask_info | ์กฐ์ฌ์ค๋ฌ์ด ๋๊ด, ๋งค์ ํ์ ์ง๋ฌธ |
|
| 94 |
+
| "ํดํน ํฐ์ง, ๋น์. ์ ๊ทผ ๊ธ์ง" | strong_neg | avoid | ๊ฐํ ๋ถ์ + ํํผ ๊ถ๊ณ |
|
| 95 |
+
| "์
๋ฐ์ดํธ ๊ณต์ง ๋์์ต๋๋ค" | neutral | info_only | ๋จ์ ์ ๋ณด ์ ๊ณต, ํ๋ ์์ |
|
| 96 |
+
|
| 97 |
+
---
|
| 98 |
+
### Citation
|
| 99 |
+
```
|
| 100 |
+
@misc{langquant2025lkbert,
|
| 101 |
+
title = {LQ-KBERT-Base: Crypto Market Korean Sentiment & Action Signal Classifier},
|
| 102 |
+
author = {LangQuant},
|
| 103 |
+
year = {2025},
|
| 104 |
+
url = {https://huggingface.co/langquant/LQ-Kbert-base}
|
| 105 |
+
}
|
| 106 |
+
```
|
| 107 |
+
---
|
| 108 |
+
### Disclaimer
|
| 109 |
+
```
|
| 110 |
+
์ด ๋ชจ๋ธ์ ํ์ ์ฐ๊ตฌ ๋ฐ ์คํ์ฉ์ผ๋ก๋ง ์ ๊ณต๋ฉ๋๋ค.
|
| 111 |
+
๋ณธ ๋ชจ๋ธ์ ์ถ๋ ฅ์ ๊ธ์ต/ํฌ์ ์๋ฌธ์ผ๋ก ๊ฐ์ฃผ๋ ์ ์์ผ๋ฉฐ,
|
| 112 |
+
๋ฐ์ํ๋ ๋ชจ๋ ๊ฒฐ๊ณผ์ ๋ํด LangQuant๋ ์ฑ
์์ ์ง์ง ์์ต๋๋ค.
|
| 113 |
+
```
|