SPTest - Intelligent Time Manager

SPTest is an advanced Android time management application with AI-powered scheduling, Chinese natural language processing, and continuous learning capabilities.

About the Project

This is a testing project developed using Claude code and Superpowers plugin. Intended to explore the workflow of Superpowers, through brainstorming, creating a git worktree, writing plans, executing plans, subagent driven developing, and completing development work after multiple rounds of TDD. There is a significant improvement in the accuracy and practicality of agent development.

🚀 Key Features

🧠 Learning System

• Continuous Improvement: Learns from user feedback (1-5 star ratings)
• Pattern Recognition: Identifies optimal time slots for different task categories
• Adaptive Scheduling: Improves task duration predictions based on historical data
• Performance Analytics: Tracks system learning progress and scheduling success rates

🗣️ Voice Recognition

• Chinese Speech-to-Text: Full voice input support with Chinese language model
• Real-time Processing: Instant voice result processing
• Permission Handling: Runtime audio permission requests
• Fallback Support: Seamlessly switches to text input if voice unavailable

📝 Chinese Natural Language Processing

• Time Expression Parsing: Supports both specific ("9点到11点") and general ("周一晚上") time expressions
• Smart Duration Calculation: Automatically calculates task duration from time ranges
• Category Detection: Identifies task types (meeting, coding, social, exercise, study)
• Priority Recognition: Understands Chinese priority keywords (紧急, 重要, 一般)
• Tag Extraction: Automatically extracts Chinese nouns as task tags

📋 Task Management

• Dual Task Lists: Separate views for Current and Completed tasks
• Long Press Delete: Task deletion with confirmation dialog
• Auto-Completion: Time-based task status updates
• Smart Scheduling: AI-powered optimal time slot suggestions

🎨 User Interface

• Professional Layout: Two ScrollView sections with proper weight distribution
• Task Cards: Formatted display with all task details
• Visual Status: Different backgrounds for task states
• Number Rating: Intuitive 1-5 number button rating system
• Toast Messages: User-friendly feedback

🛠️ Technology Stack

Android Development

• Kotlin: Primary programming language
• Jetpack Components: AppCompat, Room, WorkManager, Lifecycle
• Material Design: Modern UI/UX components
• Data Binding: Efficient view binding

AI & Machine Learning

• Custom Learning Algorithm: Bayesian-based feedback analysis
• Moving Average Updates: Smooth learning curve with 10% learning rate
• Pattern Recognition: Category-based task pattern learning
• Performance Tracking: Real-time system performance metrics

Data Processing

• JSON Storage: Efficient file-based data persistence
• Gson Serialization: Object serialization/deserialization
• User Preferences: Comprehensive settings management
• Learning Data: Historical feedback and performance tracking

NLP & Speech

• Custom Chinese Parser: Regex-based time expression parsing
• Android SpeechRecognizer: Built-in speech recognition API
• Flexible Time Understanding: Support for vague time expressions

📱 Installation & Setup

Prerequisites

• Android Studio: Hedgehog or later
• JDK: 17 or later
• Android SDK: API 34 (Android 14.0)
• Gradle: 8.7 or later

Building the Project

1. Clone the repository:

   git clone <repository-url>
   cd SPTest

2. Sync Gradle dependencies:

   • Open project in Android Studio
   • Wait for Gradle sync to complete

3. Build the APK:

   ./gradlew.bat assembleDebug

4. Run on device/emulator:

   • Connect Android device or start emulator
   • Run app from Android Studio or install APK manually

🎯 Usage Guide

Adding Tasks

Text Input:

• Type task description: "周五上午9点到11点开会"
• Press "添加任务" button
• Task is parsed and added to Current Tasks list

Voice Input:

• Press microphone button
• Speak task in Chinese: "和朋友聚餐明天晚上"
• Recognition result fills the input field
• Press "添加任务" to create task

Task Examples

• "周五上午9点到11点开会" → Meeting scheduled for Friday 9-11 AM (120 minutes)
• "紧急：完成报告大概2小时" → High priority task with 2-hour duration
• "周末游泳" → Exercise task scheduled for weekend morning
• "周一晚上和朋友聚餐" → Social task for Monday evening

Managing Tasks

Long Press Delete:

• Long press any task card
• Confirmation dialog appears
• Tap "删除" to remove task

Auto-Completion:

• Tasks automatically move to Completed list when time expires
• Expired tasks are marked as "COMPLETED"

Feedback Rating:

• Tap completed task to rate scheduling quality
• Select 1-5 stars (implemented as number buttons)
• Rating is saved and used to improve future scheduling

Smart Features

Flexible Time Handling:

• "这个周末" → This weekend based on task category
• "下周" → Next week during preferred hours
• "月底" → End of month during work hours
• "找个时间" → Tomorrow during optimal slot

AI Learning:

• System learns from each feedback rating
• Adjusts preferred times for task categories
• Improves duration predictions
• Tracks scheduling success rates

🏗️ Architecture

Project Structure

app/src/main/java/com/example/sptest/
├── MainActivity.kt              # Main UI and controller
├── TimeManagerApp.kt            # Application class
├── data/
│   ├── LearningManager.kt       # AI learning algorithm
│   ├── StorageManager.kt        # Data persistence system
│   ├── Task.kt                  # Task data model
│   ├── UserPreferences.kt       # User settings and patterns
│   └── Schedule.kt              # Schedule management
└── nlp/
    ├── ChineseTaskParser.kt     # NLP engine
    └── SimpleChineseTaskParser.kt # Simplified parser

Data Models

Task Model:

data class Task(
    val id: String,
    val title: String,
    val priority: Priority,
    val duration: Int, // minutes
    val category: String,
    val scheduledTime: TimeSlot?,
    val feedbackRating: Int,
    // ... other fields
)

User Preferences:

data class UserPreferences(
    val workHours: WorkHours,
    val taskPatterns: Map<String, TaskPattern>,
    val learningData: LearningData
)

Learning Algorithm

Feedback Analysis:

1. Stores each feedback entry with task context
2. Updates success rate based on recent 20 entries
3. Adjusts task patterns for each category
4. Learns optimal time slots based on ratings

Pattern Updates:

• Learning Rate: 10% per feedback entry
• Moving Average: Smooth adaptation to prevent overfitting
• Category-Specific: Each task type learns independently
• Time Slot Optimization: Learns preferred times from successful schedules

🔧 Configuration

Build Configuration

build.gradle.kts:

plugins {
    alias(libs.plugins.android.application)
    id("org.jetbrains.kotlin.android") version "1.9.22"
    id("org.jetbrains.kotlin.plugin.serialization") version "1.9.22"
    id("com.google.devtools.ksp") version "1.9.22-1.0.17"
}

android {
    compileSdk = 34
    defaultConfig {
        minSdk = 24
        targetSdk = 34
    }
}

AndroidManifest.xml

<uses-permission android:name="android.permission.RECORD_AUDIO" />
<uses-permission android:name="android.permission.INTERNET" />
<uses-permission android:name="android.permission.WRITE_EXTERNAL_STORAGE" />

<application
    android:name=".TimeManagerApp"
    android:theme="@style/Theme.SPTest">
    <activity android:name=".MainActivity">
        <intent-filter>
            <action android:name="android.intent.action.MAIN" />
            <category android:name="android.intent.category.LAUNCHER" />
        </intent-filter>
    </activity>
</application>

📊 Performance Metrics

The system tracks key performance indicators:

• Average Rating: User satisfaction (1-5 scale)
• Completion Rate: Task completion success percentage
• Schedule Success Rate: AI scheduling accuracy
• Total Feedback: Number of feedback entries collected

These metrics are displayed in toast messages after each feedback submission.

🌟 Future Enhancements

Potential improvements:

• Cloud Sync: Backup and sync across devices
• Calendar Integration: Export tasks to system calendar
• Reminder Notifications: Push notifications for upcoming tasks
• Advanced NLP: Support for more complex Chinese expressions
• Multi-language Support: English and other languages
• Dark Mode: Enhanced theme support

🤝 Contributing

Contributions are welcome! Please feel free to:

• Report issues
• Suggest features
• Submit pull requests
• Improve documentation

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Android Jetpack: For modern Android development components
• Gson: For JSON serialization
• Java Time API: For accurate date/time calculations
• Chinese Language Processing Community: For inspiration on NLP techniques

---

SPTest combines traditional time management with cutting-edge AI to create an intelligent, learning-enabled scheduling system that understands Chinese natural language and continuously improves based on user feedback.

For questions, issues, or feature requests, please visit the project repository.