KhmerCoreMLSMSClassifier

A Swift package for classifying Khmer SMS messages as spam or legitimate using CoreML. Built with modern Swift practices and optimized for on-device machine learning.

Features

• 🚀 Lightweight, efficient SMS classification
• 🇰🇭 Specifically designed for Khmer language messages
• 📱 Works completely offline with on-device processing
• 🔄 Handles text preprocessing for improved accuracy
• 🧠 Leverages CoreML for efficient inference
• ⚙️ Easily customizable with your own trained models

Requirements

• iOS 16.0+ / macOS 13.0+
• Swift 6.0+
• Xcode 15.0+

Installation

Swift Package Manager

Add the following line to your Package.swift dependencies:

dependencies: [
    .package(url: "https://github.com/vatanachhorn/KhmerCoreMLSMSClassifier.git", from: "1.0.0")
]

Or add it directly in Xcode:

1. Go to File > Add Package Dependencies
2. Enter the repository URL: https://github.com/vatanachhorn/KhmerCoreMLSMSClassifier.git
3. Select the version or branch you want to use

Usage

Basic Classification

import KhmerCoreMLSMSClassifier

// Initialize the classifier
let classifier = KhmerCoreMLSMSClassifier()

// Classify a Khmer message
let message = "ពេលនេះ​ លោកអ្នកទទួលបានប្រាក់ ១០០$ សូមចុចតំណរភ្ជាប់ដើម្បីទទួលយកប្រាក់ https://example.com/claim"
let category = classifier.classify(message: message)

switch category {
case .spam:
    print("This message is spam")
case .ham:
    print("This message is legitimate")
case .unknown:
    print("Could not determine message category")
}

Classification with Confidence Score

// Get classification with confidence score
let result = classifier.getDetailedClassification(message: message)
print("Category: \(result.category.rawValue), Confidence: \(result.confidence * 100)%")

// Get all confidence scores for each category
let confidences = classifier.classifyWithConfidence(message: message)
for (category, confidence) in confidences {
    print("\(category.rawValue): \(confidence * 100)%")
}

Text Preprocessing

The package includes a text preprocessor specifically designed for Khmer language:

import KhmerCoreMLSMSClassifier

let originalText = "  ពេលនេះ​ លោកអ្នកទទួលបានប្រាក់   ១០០$  សូមចុចតំណរភ្ជាប់ដើម្បីទទួលយកប្រាក់ https://example.com/claim  "
let normalizedText = TextPreprocessor.normalizeKhmerText(originalText)
print(normalizedText)
// Output: "ពេលនេះ លោកអ្នកទទួលបានប្រាក់ ១០០ សូមចុចតំណរភ្ជាប់ដើម្បីទទួលយកប្រាក់ https://example.com/claim"

iOS Example App

An iOS example application is included in the Examples directory that demonstrates all the features of the package:

Example App Features

• Basic Classification: Simple classification of Khmer SMS messages
• Detailed Classification: Visualization of classification with confidence scores
• Text Preprocessing: Demonstration of Khmer text normalization with highlighted changes

Running the Example

1. Open Examples/KhmerCoreMLSMSClassifierExample/KhmerCoreMLSMSClassifierExample.xcodeproj in Xcode
2. Select your iOS device or simulator
3. Build and run the project

For more details, see the Example App README.

Architecture

The package is organized into a modular structure:

Sources/
└── KhmerCoreMLSMSClassifier/
    ├── Core/            # Core functionality and primary classifier
    ├── Models/          # CoreML model loader and generated model classes
    ├── Utils/           # Helper utilities such as text preprocessing
    └── Resources/       # CoreML model files

Running the Console Demo

The package also includes a command-line demo application:

swift run KhmerCoreMLSMSClassifierDemo

Performance Considerations

• The classifier works entirely on-device, maintaining user privacy
• Model inference is fast and suitable for real-time classification
• Memory usage is minimal, making it suitable for mobile applications

License

This project is available under the MIT License.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add some amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Author

Created by Vatana Chhorn