ai-scam-fraud-risk-analyzer

AI Scam & Fraud Risk Analyzer

A real-time, explainable AI system that detects scam and fraud messages, assigns a risk score, and clearly explains why content may be dangerous.

---

🚀 Overview

Online scams are increasing rapidly through fake job messages, prize alerts, and malicious links.
This project helps users analyze suspicious messages before clicking by providing:

• Scam probability score
• Risk level (Low / Medium / High)
• Explainable fraud signals
• Clear safety recommendations

---

🧠 Features

• ✅ Real-time scam detection
• ✅ Hybrid AI model (ML + rule-based logic)
• ✅ Explainable risk signals
• ✅ Full-stack web application
• ✅ User-friendly interface

---

🛠️ Tech Stack

Frontend

• React (Vite)
• JavaScript
• HTML & CSS

Backend

• Flask (Python)
• Flask-CORS

AI / ML

• scikit-learn
• TF-IDF (NLP)
• Logistic Regression
• Rule-based feature extraction

---

🏗️ Architecture

React Frontend ↓ Flask REST API ↓ Hybrid AI Engine (TF-IDF + Rule Signals) ↓ Explainable Risk Output

---

⚙️ How to Run the Project Locally

1️⃣ Run Backend (Flask)

```bash cd backend python app.py

Backend will start at: http://127.0.0.1:5000

2️⃣ Run Frontend (React) cd frontend/scam-ui npm install npm run dev

Frontend will start at: http://localhost:5173

🔮 Future Enhancements

Browser extension for real-time scam detection

URL reputation and domain-age analysis

Multi-language support

Enterprise and educational security integrations

👩‍💻 Author

Bhavana Gowda GitHub: https://github.com/BhavnaGowda