abinashv29gmailcom/cybersec-validation-model-v1

Cybersecurity Data Validation Model

🛡️ Overview

This model validates user input data according to cybersecurity standards. It performs binary classification to determine if personal information fields meet security formatting requirements.

🎯 Model Purpose

• Task: Binary Text Classification (VALID/INVALID)
• Domain: Cybersecurity & Data Validation
• Use Case: Form validation, data quality checking, input sanitization

📋 Validation Rules

The model checks if input data follows these cybersecurity standards:

• firstName: Must be proper case (First letter capital, rest lowercase)

  • ✅ Valid: "John", "Alice", "Maria"
  • ❌ Invalid: "john", "ALICE", "mArIa"

• address: Each word should be properly capitalized

  • ✅ Valid: "123 Main Street", "789 Pine Road"
  • ❌ Invalid: "123 main street", "789 PINE ROAD"

• mobile: Must be exactly 10 digits

  • ✅ Valid: "9876543210"
  • ❌ Invalid: "98765", "98765432109"

• pincode: Must be exactly 6 digits

  • ✅ Valid: "560001", "400001"
  • ❌ Invalid: "560", "5600012"

🚀 Usage

Basic Usage

from transformers import pipeline

# Load the model
classifier = pipeline("text-classification", model="abinashv29gmailcom/cybersec-validation-model-v1")

# Test input
text = "firstName: John, address: 123 Main Street, mobile: 9876543210, pincode: 560001"
result = classifier(text)

print(result)
# Output: [{'label': 'VALID', 'score': 0.95}]

Batch Processing

# Multiple inputs
inputs = [
    "firstName: Alice, address: 789 Pine Road, mobile: 7654321098, pincode: 400001",
    "firstName: bob, address: main street, mobile: 98765, pincode: 123"
]

results = classifier(inputs)
for i, result in enumerate(results):
    status = "✅ VALID" if result['label'] == 'VALID' else "❌ INVALID"
    print(f"Input {i+1}: {status} (Confidence: {result['score']:.3f})")

Integration Function

def validate_user_data(firstname, address, mobile, pincode):
    input_text = f"firstName: {firstname}, address: {address}, mobile: {mobile}, pincode: {pincode}"
    result = classifier(input_text)[0]
    
    return {
        'is_valid': result['label'] == 'VALID',
        'confidence': result['score'],
        'status': result['label'],
        'input': input_text
    }

# Example usage
validation_result = validate_user_data("John", "123 Main Street", "9876543210", "560001")
print(validation_result)

📊 Examples

✅ Valid Examples

Input: "firstName: John, address: 123 Main Street, mobile: 9876543210, pincode: 560001"
Output: VALID (High Confidence)

Input: "firstName: Alice, address: 789 Pine Road, mobile: 7654321098, pincode: 400001"
Output: VALID (High Confidence)

❌ Invalid Examples

Input: "firstName: john, address: main street, mobile: 98765, pincode: 123"
Output: INVALID (Multiple violations)

Input: "firstName: MARY, address: APARTMENT 5B, mobile: 1234567890, pincode: 1234567"
Output: INVALID (Formatting issues)

🔧 Technical Details

• Base Model: distilbert-base-uncased
• Architecture: DistilBERT for Sequence Classification
• Labels: 2 classes (VALID, INVALID)
• Max Sequence Length: 128 tokens
• Framework: Transformers, PyTorch

🎯 Intended Applications

• Web Form Validation: Real-time validation of user registration forms
• Data Quality Assurance: Batch processing of existing datasets
• API Integration: RESTful services for validation endpoints
• Mobile Apps: Client-side or server-side validation
• Compliance Checking: Ensure data meets cybersecurity standards

⚠️ Limitations

• Designed for English language inputs
• Specific to the defined validation rules
• May require fine-tuning for domain-specific requirements
• Performance may vary with inputs significantly different from training examples

👨‍💻 Created By

Abinash V - Cybersecurity Data Validation System

📄 License

Apache 2.0

🔄 Version History

• v1.0: Initial model with basic cybersecurity validation rules