README.md

metadata

title: Cookie Classifier API
emoji: 🍪
colorFrom: blue
colorTo: green
sdk: docker
pinned: false
license: mit

🍪 Cookie Classifier API

FREE Serverless API for classifying web cookies into privacy categories.

🎯 What It Does

Classifies cookie names into 4 privacy categories:

• Strictly Necessary - Essential for website functionality
• Functionality - Enhance user experience
• Analytics - Track website usage and performance
• Advertising/Tracking - Marketing and ad targeting

🚀 API Endpoints

Base URL

https://aqibtahir-cookie-classifier-api.hf.space

1. Health Check

GET /

Response:

{
  "status": "online",
  "model": "Cookie Classifier - Linear Regression",
  "categories": ["Strictly Necessary", "Functionality", "Analytics", "Advertising/Tracking"]
}

2. Single Prediction

POST /predict
Content-Type: application/json

{
  "cookie_name": "_ga"
}

Response:

{
  "cookie_name": "_ga",
  "category": "Analytics",
  "class_id": 2,
  "confidence": 0.89
}

3. Batch Prediction

POST /predict/batch
Content-Type: application/json

{
  "cookie_names": ["_ga", "sessionid", "utm_campaign", "doubleclick"]
}

Response:

{
  "predictions": [
    {
      "cookie_name": "_ga",
      "category": "Analytics",
      "class_id": 2,
      "confidence": 0.89
    },
    {
      "cookie_name": "sessionid",
      "category": "Strictly Necessary",
      "class_id": 0,
      "confidence": 0.95
    },
    ...
  ]
}

💻 Usage Examples

JavaScript/Frontend

// Single prediction
async function classifyCookie(cookieName) {
  const response = await fetch('https://aqibtahir-cookie-classifier-api.hf.space/predict', {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({ cookie_name: cookieName })
  });
  
  const data = await response.json();
  console.log(data.category); // "Analytics"
  return data;
}

// Batch prediction
async function classifyManyCookies(cookieNames) {
  const response = await fetch('https://aqibtahir-cookie-classifier-api.hf.space/predict/batch', {
    method: 'POST',
    headers: {
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({ cookie_names: cookieNames })
  });
  
  const data = await response.json();
  return data.predictions;
}

// Usage
classifyCookie('_ga').then(result => {
  console.log(`${result.cookie_name} is ${result.category}`);
});

Python

import requests

# Single prediction
response = requests.post(
    'https://aqibtahir-cookie-classifier-api.hf.space/predict',
    json={'cookie_name': '_ga'}
)
print(response.json())

# Batch prediction
response = requests.post(
    'https://aqibtahir-cookie-classifier-api.hf.space/predict/batch',
    json={'cookie_names': ['_ga', 'sessionid', 'utm_campaign']}
)
print(response.json())

cURL

# Single prediction
curl -X POST "https://aqibtahir-cookie-classifier-api.hf.space/predict" \
  -H "Content-Type: application/json" \
  -d '{"cookie_name": "_ga"}'

# Batch prediction
curl -X POST "https://aqibtahir-cookie-classifier-api.hf.space/predict/batch" \
  -H "Content-Type: application/json" \
  -d '{"cookie_names": ["_ga", "sessionid", "utm_campaign"]}'

📊 Model Information

• Model: Linear Regression with TF-IDF + Name Features
• Accuracy: 90%
• Training Data: 28,671 cookie samples
• Features:
  • TF-IDF word n-grams (1-2)
  • TF-IDF char n-grams (3-5)
  • Engineered name features (length, patterns, tracker tokens, etc.)

🔒 CORS & Security

• ✅ CORS enabled for all origins (frontend-friendly)
• ✅ No authentication required for public use
• ✅ Rate-limited by Hugging Face Spaces (fair usage)

📝 Interactive Documentation

Visit the API documentation at:

https://aqibtahir-cookie-classifier-api.hf.space/docs

🏗️ Deployment

This API is deployed on Hugging Face Spaces (FREE tier) and runs 24/7.

Cold Start

• First request may take 10-30 seconds (model loading)
• Subsequent requests are fast (~100-500ms)

🤝 Support

For issues or questions:

• Model Repository: aqibtahir/cookie-classifier-lr-tfidf
• Create an issue on the Space

📄 License

MIT License - Free to use for commercial and non-commercial purposes