News Sentiment Analyzer

A Python-based sentiment analysis tool that collects live headlines from leading UK news outlets (BBC News and The Guardian) and determines sentiment using a custom rule-based NLP engine.
The tool performs headline extraction, lexical scoring, classification, and visual sentiment charting.

---

Features

• Live headline extraction from:
  • BBC News
  • The Guardian UK
• Custom-built lexicon for sentiment scoring
• Rule-based NLP pipeline (no external ML models required)
• Classification into Positive / Neutral / Negative
• Duplicate-filtering & text cleaning
• Aggregated sentiment reports per news source
• Error-handling for request failures and timeouts
• Sentiment visualization using Matplotlib (NEW)
  Automatically generates a bar chart showing sentiment distribution.

---

Technologies Used

• Python 3
• Requests
• BeautifulSoup (bs4)
• Collections (Counter)
• Matplotlib (for charts)

---

How It Works

1. Data Collection

The analyzer scrapes up to 12 <h3> headlines from BBC News and The Guardian using requests and BeautifulSoup.

2. Text Normalisation

Headlines are lower-cased, cleaned, and spacing-normalised.

3. Rule-Based Sentiment Scoring

A handcrafted lexicon assigns numeric weights:

• +1 for each positive keyword
• -1 for each negative keyword

4. Sentiment Classification

Scores are mapped into one of:

• Positive
• Neutral
• Negative

5. Reporting

The tool prints:

• Sentiment counts
• Labelled headline list
• Combined cross-source summary
• Visual sentiment chart (bar plot)

---

Example Output

Example Chart Output

Below is an example of the visual sentiment chart generated using Matplotlib:

Project Structure

• news_sentiment_analyzer.py

• README.md

  Future Improvements

• Integrate VADER or TextBlob for model-based sentiment scoring

• Expand lexicon using data-driven weighting

• Add visual sentiment charts using Matplotlib

• Build a lightweight web interface using Flask

• Add keyword trend detection and timeline analysis

• Add CSV/JSON export for sentiment reports

License

This project is released under the MIT License.
You are free to use, modify, and distribute this software with proper attribution.