Sign up or Log in to view model predictions.

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.