This project is an end-to-end Machine Learning solution built to solve a real-world problem using data-driven intelligence.
It demonstrates the complete ML lifecycle:
- Data collection
- Data preprocessing
- Feature engineering
- Model training
- Evaluation and inference
The design emphasizes clarity, scalability, and reproducibility.
- Automated data preprocessing pipeline
- Support for multiple ML models
- Clean experiment structure
- Evaluation with standard metrics
- Deployment-ready logic
This section explains how to set up and run the News Sentiment Scanner project on a local machine. The system requires Python and an active internet connection to fetch live news articles.
Ensure the following are installed:
• Python 3.9 or higher
• Stable internet connection
• Git (optional, for cloning the repository)
Check Python version:
python --version
Clone the project repository and navigate into the project directory:
git clone https://github.com/Ak47-max26/newssentimentscanner.git cd NewsSentimentScanner
Windows:
python -m venv venv venv\Scripts\activate
Linux / macOS:
python3 -m venv venv source venv/bin/activate
All required libraries are listed in requirements.txt. Install them using:
pip install -r requirements.txt
This installs dependencies for:
• News scraping (RSS, HTML parsing)
• Natural Language Processing
• Sentiment analysis
• Transformer-based models
Some NLP libraries require additional language resources. Run the following commands once:
python -m nltk.downloader punkt python -m textblob.download_corpora
Execute the main script to start sentiment analysis:
python sentiment_analysis.py
Internal workflow:
• Fetches live news from Google News RSS
• Extracts article titles
• Applies sentiment analysis using NLP models
• Displays individual results and summary insights
Raw Data
↓
Data Cleaning
↓
Feature Engineering
↓
Model Training
↓
Evaluation
↓
Prediction / Deployment