Text Classification
Transformers
Safetensors
Korean
bert
sentiment-analysis
korean
finance
finbert
Eval Results (legacy)
text-embeddings-inference
Instructions to use DataWizardd/finbert-sentiment-ko with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use DataWizardd/finbert-sentiment-ko with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="DataWizardd/finbert-sentiment-ko")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("DataWizardd/finbert-sentiment-ko") model = AutoModelForSequenceClassification.from_pretrained("DataWizardd/finbert-sentiment-ko", device_map="auto") - Notebooks
- Google Colab
- Kaggle
File size: 1,726 Bytes
156b71a 70420e5 27bafa1 156b71a 70420e5 156b71a 27bafa1 70420e5 156b71a 70420e5 156b71a 70420e5 156b71a 70420e5 27bafa1 70420e5 156b71a 27bafa1 70420e5 156b71a 70420e5 156b71a 70420e5 156b71a 27bafa1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 | ---
language: ko
datasets: naver-finance-news
tags:
- sentiment-analysis
- korean
- finance
- finbert
- transformers
license: mit
model-index:
- name: finbert-sentiment-ko
results:
- task:
name: Sentiment Analysis
type: text-classification
metrics:
- type: accuracy
value: 0.93
---
# FinBERT Sentiment Analysis (Korean, Finance Domain)
์ด ๋ชจ๋ธ์ **ํ๊ตญ์ด ํ์จ(๊ธ์ต) ๋ด์ค ์์ฝ๋ฌธ**์ ๋์์ผ๋ก ๊ฐ์ ์ ๋ถ๋ฅํ๊ธฐ ์ํด ํ์ธํ๋๋ BERT ๊ธฐ๋ฐ ๋ชจ๋ธ์
๋๋ค.
๊ฐ์ ๋ถ๋ฅ๋ ๋ค์ ์ธ ๊ฐ์ง ํด๋์ค ์ค ํ๋๋ก ์ํ๋ฉ๋๋ค:
- `0`: ๋ถ์
- `1`: ์ค๋ฆฝ
- `2`: ๊ธ์
## ๐ง ํ์ต ์ ๋ณด
- ๊ธฐ๋ฐ ๋ชจ๋ธ: [`snunlp/KR-FinBERT-SC`](https://huggingface.co/snunlp/KR-FinBERT-SC)
- ๋ฐ์ดํฐ: ์ง์ ์์งํ **๋ค์ด๋ฒ ํ์จ(๊ธ์ต) ๋ด์ค** ์์ฝ + ๊ฐ์ ์์์
๋ผ๋ฒจ๋ง
- ์ด ์ํ ์: ์ฝ 200
- Optimizer: AdamW
- Epochs: 4
- ์ต๋ ๊ธธ์ด: 128
- ํ๊ฐ ์งํ: Accuracy, F1 Score
## ๐ ์ฑ๋ฅ ํ๊ฐ
| ๊ฐ์ ํด๋์ค | Precision | Recall | F1-score | Support |
|-------------|-----------|--------|----------|---------|
| ๋ถ์ | 0.89 | 1.00 | 0.94 | 17 |
| ์ค๋ฆฝ | 1.00 | 0.82 | 0.90 | 11 |
| ๊ธ์ | 0.93 | 0.93 | 0.93 | 14 |
| **์ ํ๋** | | | **0.93** | 42 |
> ์ ์ฒด ์ ํ๋: **93%**
> Macro F1-score: **0.92**
---
## ๐ ์ฌ์ฉ ๋ฐฉ๋ฒ
```python
from transformers import pipeline
pipe = pipeline("text-classification", model="DataWizardd/finbert-sentiment-ko")
pipe("ํ์จ์ด ๊ธ๋ฑํ๋ฉฐ ์์ฅ ๋ถ์์ด ์ปค์ง๊ณ ์๋ค.")
# ์ถ๋ ฅ: [{'label': '๋ถ์ ', 'score': 0.95}]
|