Modelpublic

engibeer/prediction-sentiment-bist30

sourceHugging Facemitupdated 1y agoView on Hugging Face

0likes7downloads

Model Card

❤️ Prediction Sentiment Classifier

This model classifies the sentiment of stock prediction phrases from Turkish finance-related tweets. It’s used to assign a tone (e.g., positive or negative) to the predicted movement of a stock.

🧾 Example:

"will reach 80 TL soon" → Positive "may fall to 25 TL next week" → Negative

🧠 Model Details

—Developed by: damlakonur
—Model type: BERT fine-tuned for sequence classification
—Language(s): Turkish
—Finetuned from: bert-base-cased
—License: MIT
—Classes: LABEL_0 = Negative, LABEL_1 = Positive

🚀 How to Use

python

from transformers import pipeline

model = pipeline("text-classification", model="your-username/prediction-sentiment-bist30")

output = model("will reach 80 TL soon")
print(output)