🍽️ Food Dish Recommender System

An AI-powered dish recommendation system that suggests meals based on your preferences, mood, dietary restrictions, and allergies using content-based filtering and weighted scoring algorithms.

✨ Features

• Smart Recommendations: Content-based filtering with multi-factor scoring
• Safety First: Hard filters for allergies and dietary restrictions
• Mood-Based: Get dishes that match your current mood
• Comprehensive Database: 10,000+ real recipes from Kaggle datasets
• Explainable AI: Understand why each dish was recommended
• Cuisine Diversity: 20+ international cuisines
• Allergen Detection: Automatic allergen identification from ingredients

🧠 How It Works

The recommendation engine uses a weighted scoring system:

Final Score = (
    Preference Match × 40% +
    Mood Alignment × 30% +
    Cuisine Preference × 20% +
    Diversity Bonus × 10%
) × Allergy Filter

Scoring Components

1. Preference Match (40%): Matches your favorite ingredients
2. Mood Alignment (30%): Aligns with your current mood (comfort, energizing, light, etc.)
3. Cuisine Preference (20%): Prioritizes your favorite cuisines
4. Diversity Bonus (10%): Encourages variety by boosting less common options

Hard Filters

Before scoring, dishes are filtered by:

• ❌ Allergens (e.g., nuts, dairy, gluten)
• ❌ Dietary restrictions (vegan, vegetarian, gluten-free, etc.)
• ❌ Disliked ingredients
• ❌ Spice level tolerance
• ❌ Time constraints
• ❌ Calorie limits
• ❌ Meal type

📊 Data Sources

Uses real-world Kaggle datasets:

1. Recipe Ingredients Dataset (~40,000 recipes with cuisine classification)
2. Food Ingredients and Recipe Dataset with Images (13,582 recipes with images)
3. Ingredients with 16 Allergen Tags (10,000 USDA ingredients with allergen info)

🚀 Quick Start

Prerequisites

• Python 3.8+
• Kaggle account and API credentials

Installation

1. Clone the repository

   cd food-dish-recommender-system

2. Install dependencies

   pip install -r requirements.txt

3. Setup Kaggle credentials

   • Go to https://www.kaggle.com/settings/account
   • Scroll to "API" section
   • Click "Create New Token" to download kaggle.json
   • Place it in:
     • Windows: C:\Users\<YourUsername>\.kaggle\kaggle.json
     • Linux/Mac: ~/.kaggle/kaggle.json

4. Download datasets

   python scripts/setup_kaggle.py

   This will download ~500MB of recipe data.

5. Process datasets

   python scripts/process_data.py

   This merges, cleans, and enriches the data (takes 2-5 minutes).

6. Run the recommender!

   python src/main.py

💡 Usage Example

🍽️  FOOD DISH RECOMMENDER SYSTEM
=====================================

Tell me about your preferences!

🚨 ALLERGIES: nuts, dairy
🥗 DIETARY RESTRICTIONS: vegetarian
😊 CURRENT MOOD: 2 (energizing)
🌍 FAVORITE CUISINES: Italian, Thai
❤️  FAVORITE INGREDIENTS: pasta, basil, garlic

🔍 Finding perfect dishes for you...

🎯 YOUR PERSONALIZED RECOMMENDATIONS
=====================================

Recommendation #1
─────────────────────────────────────
🍽️  Garlic Basil Pasta (Italian)
Match Score: 87%

Why this dish:
  ✅ Contains ingredients you love (92% match)
  ✅ Perfect for your current mood (100% match)
  ✅ One of your favorite cuisines (100% match)
  ⏱️  Total time: 20 minutes
  🥗 vegetarian, dairy-free

📁 Project Structure

food-dish-recommender-system/
├── src/
│   ├── models.py                    # Data models (Dish, UserPreferences)
│   ├── recommendation_engine.py     # Core recommendation algorithm
│   ├── data_processor.py            # Dataset processing and enrichment
│   └── main.py                      # CLI application
├── scripts/
│   ├── setup_kaggle.py              # Download Kaggle datasets
│   └── process_data.py              # Process and merge datasets
├── data/
│   ├── raw/                         # Raw Kaggle datasets (gitignored)
│   └── processed/                   # Processed dish database (gitignored)
├── requirements.txt                 # Python dependencies
├── .env.example                     # Environment configuration template
└── README.md                        # This file

🎯 Algorithm Details

Content-Based Filtering

The system uses content-based filtering where dishes are recommended based on:

• Ingredient similarity to user preferences
• Cuisine type matching
• Mood tag alignment
• Allergen and dietary compatibility

Mood Tag Generation

Mood tags are automatically generated using keyword analysis:

• Comfort: Keywords like "creamy", "hearty", "warming", "traditional"
• Light: Keywords like "fresh", "salad", "crisp"
• Energizing: High-protein dishes, Asian cuisine
• Indulgent: Desserts, chocolate, rich ingredients
• Healthy: Nutritious ingredients, superfoods
• Refreshing: Cold dishes, citrus, summer foods
• Adventurous: Exotic cuisines, fusion dishes

Allergen Detection

Two-stage allergen detection:

1. Direct lookup in USDA allergen database (10,000 ingredients)
2. Keyword matching for common allergens in ingredient names

Supported allergens:

• Dairy (milk, cheese, butter, cream)
• Nuts (tree nuts, almonds, walnuts, cashews)
• Peanuts
• Eggs
• Soy
• Wheat
• Gluten
• Fish
• Shellfish
• Sesame
• Sulfites

🔧 Configuration

Copy .env.example to .env and customize:

# Recommendation Settings
MIN_MATCH_SCORE=0.3          # Minimum score threshold (0-1)
MAX_RECOMMENDATIONS=10       # Maximum results to return

# Data Paths
DATA_DIR=./data
PROCESSED_DATA_DIR=./data/processed
DB_PATH=./data/dishes.db

📈 Future Enhancements

• Machine learning model for better mood detection
• Collaborative filtering using user ratings
• Recipe image search
• Nutritional optimization
• Meal planning for multiple days
• Web UI with React
• REST API with FastAPI
• User profiles and history tracking
• Social features (share recipes, rate dishes)

🤝 Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

📄 License

MIT License - feel free to use this project for learning or commercial purposes.

🙏 Acknowledgments

• Kaggle for providing comprehensive recipe datasets
• USDA for allergen information database
• Community contributors to open food datasets

📞 Support

For issues or questions:

• Create an issue on GitHub
• Check existing issues for solutions

🍕 Enjoy Your Meal

Happy cooking! 🎉

---

Built with ❤️ for food lovers everywhere