Turkish Sentiment Analysis Model

This model is a fine-tuned version of VRLLab/TurkishBERTweet for sentiment analysis in Turkish.

Model description

The model is based on BERT and fine-tuned on a Turkish sentiment analysis dataset. It classifies text into three sentiment categories. Negative, Pozitive and Notr.

Intended uses & limitations

This model is intended for sentiment analysis of Turkish text. It should be used for research purposes or general sentiment analysis tasks in Turkish.

Training data

The model was fine-tuned on a combined dataset of labeled Turkish sentiment data.

Training procedure

The model was fine-tuned using the following hyperparameters:

learning rate: 1e-5
batch size: 8
epochs: 5 (with early stopping)
optimizer: Adam (beta1=0.9, beta2=0.999, epsilon=1e-8)

Evaluation results

Test Accuracy: 0.9325