README.md

---
license: apache-2.0
tags:
- text-classification
- cybersecurity
- data-validation
- form-validation
language:
- en
pipeline_tag: text-classification
---

# 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
```python
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
```python
# 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
```python
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