ScamShield is a web-based application designed to detect fake offers, phishing messages, and online scams using a combination of rule-based logic and Machine Learning.
The system analyzes user-input text (such as messages, links, or offers) and classifies it as Safe or High Risk, along with a confidence score. It helps users quickly identify suspicious content and avoid falling victim to cyber fraud.
- 🔍 Detects scam patterns in messages and URLs
- 🤖 Machine Learning model using Logistic Regression
- 📊 TF-IDF based text vectorization
⚠️ Risk classification: Safe 🟢 / High Risk 🔴- 📈 Confidence score for predictions
- 🌐 Simple and user-friendly web interface
-
User inputs a message or link
-
Text is processed using TF-IDF vectorization
-
Trained ML model predicts whether it's a scam
-
System displays:
- Risk level
- Confidence percentage
- Frontend: HTML, CSS
- Backend: Python (Flask)
- Machine Learning: scikit-learn (Logistic Regression, TF-IDF)
- Other Tools: Pandas, Pickle
- Detecting fake job or internship offers
- Identifying phishing messages
- Preventing online shopping scams
- Spreading cybersecurity awareness
- Model accuracy depends on dataset quality
- May not detect highly advanced scams
- Requires continuous improvement with real-world data
- Integration with real-time URL scanning APIs
- Chrome extension for instant detection
- Advanced ML/AI models for better accuracy
- Larger and real-world training datasets
With the rapid increase in cyber fraud and fake online offers, ScamShield aims to provide a simple yet effective solution to help users stay safe online by combining awareness and technology.
Cherish Sharma B.Tech CSE | SRM IST Delhi NCR