Spaces:
Sleeping
Sleeping
File size: 4,693 Bytes
89b2220 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 |
---
license: mit
title: Fraud_Detection_using_ML
sdk: docker
emoji: 🏢
colorFrom: green
colorTo: red
pinned: true
---
# Fraud Detection System
An interactive web application for real-time fraud detection using machine learning.
## Features
- **Real-time Analysis**: Instant fraud detection with probability scores
- **Interactive Interface**: User-friendly web interface with Bootstrap styling
- **Machine Learning**: Random Forest classifier trained on transaction data
- **Risk Assessment**: 5-level risk classification (Very Low to Very High)
- **Visual Results**: Progress bars and color-coded risk indicators
## Installation
1. **Install Python dependencies:**
```bash
pip install -r requirements.txt
```
2. **Run the application:**
```bash
python app.py
```
3. **Or use the batch file (Windows):**
```bash
start_app.bat
```
## Usage
1. Open your browser and go to `http://localhost:5000`
2. Fill in the transaction details:
- Personal Information (Gender, Age, Location)
- Account Information (Bank Branch, Account Type, Balance)
- Transaction Information (Amount, Type, Device, etc.)
3. Click "Analyze Transaction"
4. View the fraud probability and risk assessment
## Input Fields
### Personal Information
- **Gender**: Female or Male
- **Age**: Age of the account holder (18-100)
- **State**: Indian states (Andhra Pradesh to Ladakh)
- **City**: Major Indian cities (Mumbai, Delhi, Bangalore, etc.)
### Account Information
- **Bank Branch**: Indian bank branches (SBI, HDFC, ICICI, etc.)
- **Account Type**: Savings, Current, Credit Card, Fixed Deposit, Salary, Joint
- **Account Balance**: Current account balance in Indian Rupees (₹)
### Transaction Information
- **Transaction Amount**: Amount being transacted in Indian Rupees (₹)
- **Transaction Type**: ATM Withdrawal, Bank Deposit, Fund Transfer, UPI Payment, etc.
- **Transaction Device**: ATM, POS, Mobile Apps, UPI platforms, Internet Banking, etc.
- **Transaction Date**: Day of month (1-31)
- **Transaction Time**: Hourly time slots (24-hour format)
- **Currency**: INR (default), USD, EUR, GBP, AED, SGD, CAD, AUD
## API Endpoints
### GET /
Returns the main web interface
### POST /predict
**Request Body:**
```json
{
"gender": "1",
"age": "35",
"state": "15",
"city": "127",
"bank_branch": "127",
"account_type": "2",
"transaction_date": "22",
"transaction_time": "52151",
"transaction_amount": "32415.45",
"transaction_type": "3",
"account_balance": "74557.27",
"transaction_device": "17",
"transaction_currency": "0"
}
```
**Response:**
```json
{
"prediction": 0,
"fraud_probability": 15.75,
"risk_level": "Very Low",
"risk_color": "#28a745",
"message": "Legitimate Transaction"
}
```
## Risk Levels
- **Very Low** (0-20%): Green (#28a745)
- **Low** (20-40%): Teal (#20c997)
- **Medium** (40-60%): Yellow (#ffc107)
- **High** (60-80%): Orange (#fd7e14)
- **Very High** (80-100%): Red (#dc3545)
## Testing
Run the API test script:
```bash
python test_api.py
```
## Files Structure
```
fraud_flask_app/
├── app.py # Main Flask application
├── model.pkl # Trained Random Forest model
├── encoders.pkl # Label encoders for categorical features
├── requirements.txt # Python dependencies
├── start_app.bat # Windows startup script
├── test_api.py # API testing script
├── templates/
│ └── index.html # Web interface template
├── static/
│ └── style.css # Custom styling
└── README.md # This file
```
## Model Information
- **Algorithm**: Random Forest Classifier
- **Features**: 13 input features
- **Training Data**: Bank transaction fraud dataset
- **Performance**: Optimized for fraud detection accuracy
## Technologies Used
- **Backend**: Flask (Python)
- **Frontend**: HTML5, CSS3, JavaScript, Bootstrap 5
- **Machine Learning**: scikit-learn, pandas, numpy
- **Icons**: Font Awesome
## Browser Support
- Chrome/Chromium (recommended)
- Firefox
- Safari
- Edge
## Troubleshooting
### Model Loading Issues
- Ensure `model.pkl` and `encoders.pkl` are in the same directory as `app.py`
- Check Python package versions match requirements.txt
### Connection Issues
- Verify Flask is running on port 5000
- Check firewall settings
- Ensure no other application is using port 5000
### Prediction Errors
- Validate all input fields are filled correctly
- Check input value ranges match the specified limits
- Ensure numeric fields contain valid numbers
## License
This project is for educational and demonstration purposes. |