Sentiscope

Sentiscope is a powerful Python package for analyzing sentiment and extracting entities from financial news articles. It leverages multiple state-of-the-art sentiment analysis models and named entity recognition techniques to provide comprehensive insights into financial news.

Features

• Fetch financial news articles for specific sectors using NewsAPI
• Perform sentiment analysis using multiple models:
  • FinBERT
  • VADER
  • ESG-BERT
  • FinBERT-Tone
  • Flair
• Extract named entities using:
  • NLTK
  • Flair

Installation

pip install sentiscope

Quick Start

from sentiscope import SentimentAnalyzer, NewsFetcher, EntityExtractor

# Initialize components
sentiment_analyzer = SentimentAnalyzer()
news_fetcher = NewsFetcher(api_key="your_newsapi_key_here")
entity_extractor = EntityExtractor()

# Fetch news
news = news_fetcher.fetch_financial_news("Banking")

# Analyze sentiment and extract entities
for article in news:
    finbert_sentiment = sentiment_analyzer.analyze_sentiment_finbert(article['description'])
    vader_sentiment = sentiment_analyzer.analyze_sentiment_vader(article['description'])
    entities = entity_extractor.extract_entities_flair(article['description'])
    
    print(f"Title: {article['title']}")
    print(f"FinBERT Sentiment: {finbert_sentiment}")
    print(f"VADER Sentiment: {vader_sentiment}")
    print(f"Entities: {entities}")
    print("---")

Detailed Usage

Sentiment Analysis

Sentiscope offers multiple sentiment analysis models:

text = "The company reported strong earnings, beating market expectations."

finbert_sentiment = sentiment_analyzer.analyze_sentiment_finbert(text)
vader_sentiment = sentiment_analyzer.analyze_sentiment_vader(text)
esgbert_sentiment = sentiment_analyzer.analyze_sentiment_esgbert(text)
finbert_tone_sentiment = sentiment_analyzer.analyze_sentiment_finbert_tone(text)
flair_sentiment = sentiment_analyzer.analyze_sentiment_flair(text)

News Fetching

Fetch financial news for specific sectors:

news = news_fetcher.fetch_financial_news("Technology")

Models

Sentiscope uses the following pre-trained models:

• FinBERT: Fine-tuned BERT for financial sentiment analysis
• ESG-BERT: BERT model for environmental, social, and governance (ESG) sentiment
• FinBERT-Tone: BERT model for financial sentiment tone analysis
• Flair: General-purpose sentiment analysis and named entity recognition
• VADER: Rule-based sentiment analysis tool

Dependencies

• transformers
• torch
• nltk
• flair
• requests

Note

This package requires an API key from NewsAPI to fetch financial news articles. You can obtain a key by registering at https://newsapi.org.

License

This project is licensed under the MIT License.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

Support

If you encounter any issues or have questions, please file an issue on the GitHub repository.

Acknowledgements

Sentiscope is built on top of several open-source projects and pre-trained models. We are grateful to the developers and researchers who have made their work available to the community.