🧬 Decision DNA

AI Governance & Monitoring System for Credit Risk Models

Decision DNA is an AI governance and monitoring platform designed to supervise machine learning models used in credit risk decision systems.

The system helps detect model drift, operational risks, and security threats, while maintaining transparent and auditable AI decision pipelines.

---

🏛️ System Architecture

User
  ↓
React Governance Dashboard
  ↓
Express API Layer
  ↓
Model Monitoring Engine
  ↓
Drift Detection (PSI / KL Divergence)
  ↓
Security Layer
  ↓
Audit Logging System

Decision DNA provides real-time insights into model behavior, prediction monitoring, and governance alerts.

---

✨ Features

📊 Live Governance Dashboard

Real-time monitoring of:

• model health
• prediction activity
• system alerts
• governance metrics

---

🔍 Model Drift Detection

Detects distribution shifts using statistical techniques:

• Population Stability Index (PSI)
• KL Divergence

This helps identify when production data deviates from training data.

---

🛡️ Security Monitoring

Provides monitoring mechanisms to detect:

• abnormal model behavior
• suspicious system activity
• operational failures

Security events are logged for auditing.

---

🤖 Risk Prediction Interface

Interactive interface to test credit risk predictions using trained models.

---

📋 Audit Logging

All governance events and system activities are stored as timestamped audit logs.

---

🔄 Recovery & Reboot System

Administrative tools allow safe restart or recovery of model services when anomalies occur.

---

🏗️ Tech Stack

Layer	Technology
Frontend	React + TypeScript + Tailwind CSS
Backend	Node.js + Express
Build Tool	Vite
ML	Python + scikit-learn
Validation	Zod
Storage	IndexedDB / local logs

---

📁 Project Structure

Decision-DNA/
│
├── screenshots/          # UI screenshots for README
│
├── src/
│   ├── api/              # API client layer
│   ├── config/           # configuration files
│   ├── drift/            # drift detection logic
│   ├── pages/            # React UI pages
│   ├── services/         # backend services
│
│   ├── App.tsx
│   ├── ai_report.ts
│   ├── constants.tsx
│   ├── index.css
│   └── index.tsx
│
├── ml/                   # ML training pipeline
├── scripts/              # automation scripts
├── security/             # security monitoring modules
├── tests/                # testing files
│
├── server.ts             # Express backend server
│
├── index.html
├── package.json
├── tsconfig.json
├── vite.config.ts
├── tailwind.config.js
├── postcss.config.js
│
├── .env.example          # environment variables template
├── .gitignore
├── SECURITY.md
└── README.md

---

📊 Dataset

The system uses a synthetic credit risk dataset for training and demonstration.

Example features:

Feature	Description
creditScore	Credit score range (300–850)
income	Annual applicant income
debtRatio	Debt-to-income ratio
loanAmount	Requested loan amount
decision	Loan approval outcome

---

🚀 Quick Start

1️⃣ Clone the repository

git clone https://github.com/Harshit-it25/Decision-DNA.git
cd Decision-DNA

---

2️⃣ Install dependencies

npm install

---

3️⃣ Configure environment variables

Create .env.local:

API_KEY=your_api_key

---

4️⃣ Start the development server

npm run dev

This launches:

• Express backend
• React dashboard

---

5️⃣ Open the application

http://localhost:5173

---

📦 Available Scripts

Command	Description
npm run dev	Start development server
npm run build	Build production bundle
npm run preview	Preview production build
npm run lint	Type checking

---

🔐 Security Policy

Security guidelines and reporting instructions are available in:

SECURITY.md

---

📈 Future Improvements

Possible enhancements include:

• automated model retraining
• real-time drift alerts
• anomaly detection
• explainable AI modules
• cloud deployment

---

👨‍💻 Author

Harshit Ranbhare

AI / Machine Learning Student Btech IT Student

---

📄 License

MIT License © 2024 Harshit-it25

---

⭐ If you found this project interesting, consider starring the repository.