An intelligent spam classification system that uses machine learning to identify spam messages with high accuracy. The system features a modern dark-themed UI and provides detailed analysis of message characteristics.
🚀 Check out the live version of Email SMS Spam Classifier Model! 🚀
- Real-time spam detection
- Confidence score visualization
- Suspicious feature detection
- Dark mode interface
- Detailed text analysis
- 97.67% accuracy on test data
- Clone the repository:
git clone https://github.com/yourusername/spam-classifier.git
cd spam-classifier- Install dependencies:
pip install -r requirements.txt- Run the application:
streamlit run app.py- Lottery Scam:
CONGRATULATIONS! You've won £1,000,000 in the UK lottery! To claim your prize, contact us at +44-XXX-XXXX or reply NOW!
- Prize Winner:
You are the lucky winner of 2 lakh rupees! Contact immediately to claim your cash prize! Urgent - respond within 24hrs!!!
- Banking Scam:
Dear Customer, your account will be suspended! Update your KYC by clicking http://fakebank.com immediately!
- Marketing Spam:
FREE GIFT! Buy one get THREE free! Limited time offer - 90% OFF on all items! Shop now at www.fakeshop.com
- Investment Scam:
INVEST NOW! 1000% guaranteed returns in crypto! Don't miss this opportunity. Contact our expert: +1-XXX-XXXX
- Regular Meeting:
Hi Team, reminder about our weekly meeting tomorrow at 10 AM. Please prepare your updates.
- Friend's Message:
Hey! Are we still on for dinner tonight at 7? Let me know if you need directions to the restaurant.
- Delivery Update:
Your package has been delivered. Thank you for shopping with us!
- Birthday Wish:
Happy Birthday! Hope you have a wonderful day filled with joy and laughter. 🎂
- Work Related:
Please review the attached document and send your feedback by end of day.
The classifier looks for several suspicious patterns:
- Multiple exclamation marks (!!!)
- Currency symbols (₹, $, £, €)
- Excessive uppercase text
- Multiple numbers
- Common spam keywords (free, win, prize, urgent)
- URLs and phone numbers
- Unusual punctuation patterns
- Training Accuracy: 99.17%
- Testing Accuracy: 97.67%
- Cross-validation Score: 98.50%
Feel free to contribute to this project by:
- Forking the repository
- Creating a new branch
- Making your changes
- Submitting a pull request
- Dataset source: UCI Machine Learning Repository
- NLTK for text processing
- Streamlit for the web interface
- scikit-learn for machine learning