SpamVision is an AI-based email classification system that analyzes incoming emails and classifies them as Spam, Phishing, Promotional, or Safe using NLP + Machine Learning models.
It supports real-time monitoring and works with multiple email inboxes.
SpamVision intelligently categorizes emails into:
- 📮 Safe – Normal, legitimate emails
- 🚫 Spam – Junk mail or bulk advertisements
- 🎣 Phishing – Fraud attempts to steal data
- 🛍️ Promotional – Marketing & offer-based emails
The model uses a vectorizer + ML classifier trained on email datasets (available in the training/ folder).
- ✔ Machine Learning–based email classification
- ✔ Real-time monitoring using Python script
- ✔ Lightweight & fast
- ✔ Works on any IMAP-supported inbox
- ✔ Training scripts included
- ✔ Notification support (desktop alerts)
- Python 3.10+
- Libraries: scikit-learn, numpy, pandas, nltk
- Model Type: ML classifier (trained model
.pkl) - Techniques: Bag-of-Words / TF-IDF, preprocessing, tokenization
| File/Folder | Description |
|---|---|
spamvision_model.pkl |
Trained machine learning email classifier |
vectorizer.pkl |
Vectorizer used to preprocess email text |
spamvision_monitor.py |
Real-time email monitoring script |
training/ |
Training scripts & datasets used for model development |
requirements.txt |
List of Python packages |
sample_emails.csv |
Sample training/validation dataset |
NotificationDetected.jpg |
Screenshot of desktop notification |
ZohoMailbox.png |
Screenshot of email inbox |
ProjectDemo.mp4 |
Demo video of the system running |
.gitattributes |
Git attribute configuration |
spamvision/
├── training/
├── spamvision_monitor.py
├── spamvision_model.pkl
├── vectorizer.pkl
├── sample_emails.csv
├── requirements.txt
├── ProjectDemo.mp4
├── ZohoMailbox.png
├── NotificationDetected.jpg
└── README.md
git clone https://github.com/vaishnaviteja05/SmartVision-A-smart-email-classifier.git
cd spam-visionpip install -r requirements.txtpython spamvision_monitor.py