🛒 Product Recommendation System

An intelligent product recommendation system using semantic similarity and sentence transformers. Analyzes conversation context to recommend relevant products across multiple categories.

✨ Features

• 🔍 Semantic Understanding - Uses sentence transformers for context analysis
• 🛍️ Multi-Category Products - Technology, entertainment, fashion, health, food
• 📊 Cosine Similarity Matching - Accurate product relevance scoring
• 💬 Conversation Analysis - Extracts intent from natural language
• 🎯 Keyword Matching - Hybrid approach for better accuracy
• 📈 SamSum Integration - Text summarization capabilities

🚀 Getting Started

Prerequisites

• Python 3.8+
• ~500MB for sentence transformer model

Installation

1. Clone the repository

   git clone https://github.com/yourusername/product-recommendation.git
   cd product-recommendation

2. Create virtual environment

   python -m venv venv
   source venv/bin/activate  # Windows: venv\Scripts\activate

3. Install dependencies

   pip install -r requirements.txt

4. Run the system

   python product_recommendation_system.py

📁 Project Structure

product-recommendation/
├── .github/
│   ├── workflows/
│   │   └── ci.yml
│   ├── ISSUE_TEMPLATE.md
│   └── PULL_REQUEST_TEMPLATE.md
├── src/
│   ├── product_recommendation_system.py  # Main recommendation engine
│   └── SamSum.py                          # Text summarization
├── data/
│   └── samsum_data/                       # Training data
├── .gitignore
├── .env.example
├── requirements.txt
├── LICENSE
├── CONTRIBUTING.md
├── CHANGELOG.md
└── README.md

🛍️ Product Categories

Category	Example Products
Technology	Smartphones, Apps
Entertainment	Netflix, Concerts
Fashion	Clothing, Accessories
Health	Supplements, Ergonomic furniture
Food	Meal delivery services
Books	E-readers, Book subscriptions

🔧 How It Works

1. Input Processing - Receives conversation or query text
2. Embedding Generation - Creates semantic embeddings using all-MiniLM-L6-v2
3. Similarity Calculation - Computes cosine similarity with product database
4. Keyword Enhancement - Boosts matches based on keyword overlap
5. Ranking - Returns top-N most relevant products

💻 Usage Example

from product_recommendation_system import ProductRecommendationSystem

recommender = ProductRecommendationSystem()

# Get recommendations from conversation
conversation = "I've been feeling tired lately and need more energy for work."
recommendations = recommender.get_recommendations(conversation)

for product in recommendations:
    print(f"{product['name']}: {product['score']:.2f}")

🤝 Contributing

Contributions are welcome! Please read our Contributing Guide.

📄 License

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

👤 Author

Your Name

• GitHub: @yourusername

---

Made with ❤️ and Sentence Transformers