Anutri03/url-phishing-detection-api

URL Phishing Detection API

A machine learning API that detects phishing URLs using LightGBM. This API can be integrated into websites to provide real-time URL security analysis.

Features

• —Single URL Prediction: Analyze individual URLs for phishing attempts
• —Batch Processing: Process multiple URLs at once (up to 100)
• —Feature Extraction: Get detailed URL features for analysis
• —High Accuracy: Trained on balanced dataset with LightGBM

API Endpoints

Health Check

GET /

Returns API status and available endpoints.

Single URL Prediction

POST /predict
Content-Type: application/json

{
  "url": "https://example.com"
}

Response:

{
  "url": "https://example.com",
  "prediction": "safe",
  "probability": 0.15,
  "confidence": 0.85,
  "status": "success"
}

Batch URL Prediction

POST /predict_batch
Content-Type: application/json

{
  "urls": [
    "https://example.com",
    "http://suspicious-site.com"
  ]
}

Feature Extraction

POST /features
Content-Type: application/json

{
  "url": "https://example.com"
}

Integration Example

JavaScript/Node.js

const response = await fetch('https://your-space-name.hf.space/predict', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
  },
  body: JSON.stringify({
    url: 'https://example.com'
  })
});

const result = await response.json();
console.log(result.prediction); // "safe" or "phishing"

Python

import requests

response = requests.post('https://your-space-name.hf.space/predict', 
                        json={'url': 'https://example.com'})
result = response.json()
print(result['prediction'])  # "safe" or "phishing"

cURL

curl -X POST https://your-space-name.hf.space/predict \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com"}'

Model Details

• —Algorithm: LightGBM (Gradient Boosting)
• —Features: 15 lexical features extracted from URLs
• —Training: Balanced dataset with early stopping
• —Performance: Optimized for speed and accuracy

Security Features

• —Input validation and sanitization
• —Error handling for malformed URLs
• —Rate limiting through Hugging Face infrastructure
• —CORS enabled for web integration

Usage for dhaal.io

This API is designed to be integrated into your dhaal.io website. Simply make HTTP requests to the deployed endpoint to get real-time phishing detection for any URL.

API Base URL: https://your-space-name.hf.space

Replace your-space-name with your actual Hugging Face Space name after deployment.