An intelligent AI-powered cybersecurity platform that combines Financial Fraud Detection, Network Intrusion Detection, Real-Time Threat Monitoring, Risk Scoring, and Interactive Analytics into a unified dashboard.
SENTINEL is a full-stack AI-powered cybersecurity platform designed to identify suspicious activities across multiple security domains.
The system integrates modern Machine Learning techniques with a scalable FastAPI backend and a React-based dashboard to provide real-time threat detection, risk analysis, explainable alerts, and security analytics.
Unlike traditional monitoring systems that only generate alerts, SENTINEL provides contextual insights, risk scoring, and visual analytics to help security analysts understand and respond to threats more effectively.
The primary objectives of SENTINEL are:
- 🔍 Detect anomalous activities in real time
- 💳 Identify fraudulent financial transactions
- 🌐 Detect malicious network intrusions
- 📊 Generate intelligent risk scores
- 🧠 Provide Explainable AI insights
- 📈 Visualize live security events
- 📄 Generate downloadable reports
- 🔐 Secure administrator and user access
- ⚡ Deliver a responsive real-time dashboard
- Real-time anomaly detection
- AI-powered fraud detection
- Network intrusion detection
- Risk score generation
- Threat categorization
- Live event monitoring
- Admin Dashboard
- User Dashboard
- Analytics Page
- Reports Page
- Alerts Page
- System Settings
- Real-Time Statistics
- Risk Visualization
- Secure Login
- User Authentication
- Role-Based Access Control
- Protected Routes
- Session Management
- Risk Gauge
- Threat Distribution
- Security Metrics
- Live Alert Feed
- Performance Statistics
- Report Generation
- REST APIs
- Alert Management
- Authentication APIs
- Risk Engine
- Event Processing
- Report Services
┌───────────────────────────┐
│ Web Browser │
└─────────────┬─────────────┘
│
▼
┌─────────────────────────────────┐
│ React + TypeScript │
│ Vite + Tailwind CSS │
└─────────────┬───────────────────┘
│ REST API
▼
┌─────────────────────────────────┐
│ FastAPI Backend │
├─────────────────────────────────┤
│ Authentication │
│ Alert Management │
│ Risk Score Engine │
│ Event Processing │
│ Report Generation │
└─────────────┬───────────────────┘
│
┌───────────────────────┼────────────────────────┐
▼ ▼ ▼
┌────────────────┐ ┌────────────────┐ ┌────────────────┐
│ Fraud Detection│ │ Intrusion IDS │ │ Risk Engine │
│ Machine Learning│ │ ML Pipeline │ │ AI Analytics │
└────────────────┘ └────────────────┘ └────────────────┘
│
▼
Security Intelligence
SENTINEL/
│
├── backend/
│ ├── app/
│ │ ├── api/
│ │ │ └── routers/
│ │ ├── config/
│ │ ├── database/
│ │ ├── middleware/
│ │ ├── models/
│ │ ├── schemas/
│ │ └── services/
│ │
│ └── requirements.txt
│
├── frontend/
│ ├── public/
│ ├── src/
│ │ ├── assets/
│ │ ├── components/
│ │ ├── context/
│ │ ├── hooks/
│ │ ├── pages/
│ │ ├── services/
│ │ ├── types/
│ │ └── utils/
│ │
│ ├── package.json
│ └── vite.config.ts
│
├── datasets/
│
├── PROJECT_REQUIREMENTS.md
├── README.md
└── .env.example
| Category | Technologies |
|---|---|
| Frontend | React, TypeScript, Vite |
| Styling | Tailwind CSS |
| Backend | FastAPI, Python |
| API | REST API |
| Authentication | JWT Authentication |
| Machine Learning | Scikit-learn |
| Data Processing | Pandas, NumPy |
| Database | SQLite / PostgreSQL |
| Validation | Pydantic |
| Charts | Custom Dashboard Widgets |
| Version Control | Git & GitHub |
The Threat Detection Engine continuously analyzes incoming events to identify suspicious activities using AI-driven anomaly detection techniques.
- Real-time anomaly detection
- Threat classification
- Event prioritization
- Confidence scoring
- Alert generation
Detects fraudulent financial transactions using machine learning models trained on transaction patterns.
- Credit card fraud detection
- Transaction risk scoring
- Behavioral analysis
- Fraud probability estimation
- High-risk transaction alerts
Monitors network traffic to detect malicious behavior and unauthorized access attempts.
- DoS / DDoS attacks
- Port scanning
- Brute-force attacks
- Malware communication
- Suspicious traffic patterns
- Network anomalies
Every detected event receives an intelligent risk score between 0–100 based on multiple parameters.
| Score | Severity |
|---|---|
| 0–20 | Low |
| 21–40 | Moderate |
| 41–60 | Medium |
| 61–80 | High |
| 81–100 | Critical |
The dashboard provides real-time visibility into system health and detected threats.
- Live alerts
- Threat statistics
- Security overview
- Risk visualization
- Historical trends
- Reports
- User activity
- Performance metrics
- JWT-based Authentication
- Secure Login System
- Role-Based Access Control (RBAC)
- Protected Routes
- Session Management
- User Access Validation
- Live Alert Monitoring
- Threat Categorization
- Risk-Based Prioritization
- Alert Status Tracking
- Alert History
- Event Logging
- Interactive Dashboard
- Real-Time Metrics
- Threat Statistics
- Risk Distribution
- Recent Alerts
- Activity Monitoring
- Security Overview
- Performance Charts
- Security Reports
- Alert Reports
- Threat Summary
- CSV Export
- PDF Report Generation
- Historical Analysis
- RESTful API
- FastAPI Framework
- Authentication Services
- Alert Services
- Risk Engine
- Event Processing
- Report Services
git clone https://github.com/Chirag04-bit/SENTINAL.git
cd SENTINALNavigate to the backend directory.
cd backendCreate a virtual environment.
python -m venv venv
venv\Scripts\activatepython3 -m venv venv
source venv/bin/activateInstall dependencies.
pip install -r requirements.txtRun the backend server.
uvicorn app.main:app --reloadBackend URL
http://127.0.0.1:8000
Swagger Documentation
http://127.0.0.1:8000/docs
ReDoc
http://127.0.0.1:8000/redoc
Navigate to frontend.
cd frontendInstall packages.
npm installRun the development server.
npm run devFrontend URL
http://localhost:5173
Create a .env file inside the backend directory.
DATABASE_URL=sqlite:///./sentinel.db
SECRET_KEY=your-secret-key
ALGORITHM=HS256
ACCESS_TOKEN_EXPIRE_MINUTES=30
DEBUG=TrueExample:
backend/
.env
After running the backend, visit:
Swagger UI
http://127.0.0.1:8000/docs
ReDoc
http://127.0.0.1:8000/redoc
Current API services include:
- Authentication
- User Management
- Alert Management
- Risk Analysis
- Reports
- Dashboard Analytics
POST /login
POST /register
POST /refresh-token
GET /alerts
GET /alerts/{id}
POST /alerts
PUT /alerts/{id}
DELETE /alerts/{id}
GET /reports
POST /reports
GET /reports/download
GET /analytics
GET /statistics
GET /risk-score
Clone Repository
│
▼
Install Dependencies
│
▼
Configure Environment
│
▼
Run Backend
│
▼
Run Frontend
│
▼
Access Dashboard
│
▼
Start Development
Incoming Event
│
▼
Threat Detection
│
▼
Risk Analysis
│
▼
Alert Generation
│
▼
Dashboard Update
│
▼
Report Generation
SENTINEL leverages Artificial Intelligence and Machine Learning to identify suspicious activities across multiple cybersecurity domains.
The platform is designed to support both supervised and unsupervised learning techniques for anomaly detection and fraud analysis.
Raw Data
│
▼
Data Cleaning
│
▼
Feature Engineering
│
▼
Model Training
│
▼
Model Evaluation
│
▼
Risk Scoring
│
▼
Threat Classification
│
▼
Dashboard Visualization
- Fraud Detection
- Network Intrusion Detection
- Anomaly Detection
- Risk Prediction
- Threat Classification
- Pattern Recognition
- Security Analytics
- Explainable AI (SHAP)
- Explainable AI (LIME)
- Deep Learning Models
- AutoML Pipeline
- Online Learning
- Real-Time Model Retraining
The project has been designed to work with multiple public cybersecurity datasets.
| Dataset | Purpose |
|---|---|
| Credit Card Fraud Detection | Financial Fraud Detection |
| PaySim Financial Dataset | Transaction Fraud Analysis |
| NSL-KDD | Network Intrusion Detection |
| UNSW-NB15 | Modern Network Attack Detection |
Note: Large datasets are intentionally excluded from the GitHub repository. Download them separately and place them inside the
datasets/directory.
Example:
datasets/
├── creditcard.csv
├── PS_20174392719_1491204439457_log.csv
├── KDDTrain+.txt
├── KDDTest+.txt
├── UNSW_NB15_training-set.csv
└── UNSW_NB15_testing-set.csv
Add screenshots after completing the UI.
docs/screenshots/login.png
docs/screenshots/dashboard.png
docs/screenshots/analytics.png
docs/screenshots/alerts.png
docs/screenshots/reports.png
| Module | Status |
|---|---|
| Frontend UI | ✅ Completed |
| Backend APIs | ✅ Completed |
| Authentication | ✅ Completed |
| Dashboard | ✅ Completed |
| Alert System | ✅ Completed |
| Analytics | ✅ Completed |
| Reports | ✅ Completed |
| Risk Engine | 🚧 In Progress |
| Machine Learning Integration | 🚧 In Progress |
| Database Integration | 🚧 In Progress |
| Explainable AI | 📅 Planned |
| Deployment | 📅 Planned |
- React Frontend
- FastAPI Backend
- Authentication
- Dashboard
- Alerts
- Reports
- Analytics
- Risk Visualization
- PostgreSQL Integration
- Redis Cache
- Docker Support
- Logging System
- Unit Testing
- API Rate Limiting
- Real-Time WebSockets
- Email Notifications
- SMS Alerts
- AI Explainability
- Multi-Factor Authentication
- Cloud Deployment
- Kubernetes Deployment
- Multi-Tenant Support
- SOC Dashboard
- Threat Intelligence Feed
- SIEM Integration
- Mobile Application
Contributions are welcome!
If you would like to improve SENTINEL:
- Fork the repository
- Create a new feature branch
git checkout -b feature/your-feature
- Commit your changes
git commit -m "Add new feature"
- Push your branch
git push origin feature/your-feature
- Open a Pull Request
Backend
pytest
Frontend
npm test
This project is licensed under the MIT License.
Feel free to use, modify, and distribute this project for educational and research purposes.
B.Tech – Computer Science & Engineering (AI)
Institute of Engineering & Management (IEM), Kolkata
- GitHub: https://github.com/Chirag04-bit
- LinkedIn: https://www.linkedin.com/in/chirag-sharma-5a79b4322/
Special thanks to the open-source community and the creators of:
- FastAPI
- React
- TypeScript
- Vite
- Tailwind CSS
- Scikit-learn
- NumPy
- Pandas
- PostgreSQL
- Python
If you found this project useful, consider giving it a ⭐ on GitHub.
It helps the project reach more developers and motivates future improvements.
Built with ❤️ by Chirag Sharma
⭐ Don't forget to Star this repository! ⭐
Coming Soon
The application demo will showcase:
- User Authentication
- Admin Dashboard
- Threat Monitoring
- Fraud Detection
- Risk Score Engine
- Analytics Dashboard
- Report Generation
| Login | Dashboard |
|---|---|
 |
 |
| Alerts | Analytics |
|---|---|
 |
 |
| Reports |
|---|
 |
| Metric | Target |
|---|---|
| API Response Time | < 200 ms |
| Dashboard Load Time | < 2 sec |
| Authentication | < 500 ms |
| Risk Score Generation | < 1 sec |
| Fraud Detection Accuracy | > 95% |
| Intrusion Detection Accuracy | > 95% |
The current version is intended for educational, research, and portfolio purposes. Some production-grade features such as deployment, monitoring, and advanced security are planned for future releases.
The project supports public cybersecurity datasets including:
- Credit Card Fraud Detection
- PaySim
- NSL-KDD
- UNSW-NB15
Yes. The platform is designed to integrate machine learning models for fraud detection, anomaly detection, and network intrusion detection.
Absolutely! Contributions, bug reports, and feature requests are welcome.
- Machine Learning models are under active development.
- Database integration is being expanded.
- WebSocket support is planned.
- Cloud deployment is not yet available.
- AI-powered threat intelligence
- Explainable AI dashboards
- Live WebSocket streaming
- Docker support
- Kubernetes deployment
- Redis caching
- Multi-factor authentication
- SIEM integration
- Email & SMS alerts
- Mobile application
- Cloud-native deployment
- Multi-tenant architecture
- Initial project structure
- React frontend
- FastAPI backend
- Authentication module
- Dashboard UI
- Alert management
- Analytics pages
- Report module
If you discover a security vulnerability, please create a private issue or contact the maintainer before publicly disclosing it.
If you use this project in your research or academic work, please cite it appropriately.
Chirag Sharma.
SENTINEL: Smart Emergency Network for Threat Intelligence &
Network Evaluation Layer.
GitHub Repository.
2026.
If you like this project:
⭐ Star the repository
🍴 Fork it
🛠️ Contribute
📢 Share it with others