distilroberta-financial-news-sentiment-v2
The DistilRoberta-financial-sentiment-v2 is a fast, case-sensitive model fine-tuned on financial news for high-accuracy sentiment analysis.
1
Pre-tax loss totaled euro 0.3 million, compared to a loss of euro 2.2 million in the first quarter of 2005 .
Notes
Introduction
DistilRoberta-financial-sentiment-v2 model, a fine-tuned version of the DistilRoBERTa model on the financial_phrasebank dataset for sentiment analysis of financial news.
DistilRoberta-financial-sentiment-v2 Model
The DistilRoberta-financial-sentiment-v2 model is based on the DistilRoBERTa model, which is a distilled version of the RoBERTa-base model. It consists of totaling 82 million parameters. The model is case-sensitive, distinguishing between English and English. It is twice as fast as the RoBERTa-base model.
Use Cases
The model is trained to analyze sentiment in financial news text. Potential use cases include:
- Sentiment analysis of financial articles and reports
- Automated trading strategies based on sentiment signals
- Market sentiment analysis for investment decision-making
Evaluation
Base Model Performance
- Loss: 0.1116
- Accuracy: 0.9923
Training Data
The model was trained on a polar sentiment dataset containing 4840 sentences from English language financial news. The dataset is categorized by sentiment and annotated by 5-8 annotators.
Advantages
- Fast inference time compared to larger models like RoBERTa-base.
- Trained specifically for sentiment analysis in financial news, potentially providing more accurate results in this domain.
- Case-sensitive, which can capture subtle differences in sentiment based on language nuances.