E-commerce Product Recommendation System

A Java-based product recommendation system that uses OpenAI's GPT API to provide personalized product recommendations and sentiment analysis for customer reviews.

Features

• Product recommendations based on user input
• Multi-profile recommendations
• Sentiment analysis for product reviews
• Token usage tracking and cost calculation
• CSV file processing support

Technologies

• Java 21
• Maven
• OpenAI GPT API (simple-openai)
• JTokkit for token counting

Project Structure

src/
├── main/
│   ├── java/
│   │   └── dev/raniery/ecommerce/
│   │       ├── details/                # System configurations
│   │       ├── functions/              # Core functionality
│   │       └── IntegrationTest         # OpenAI API integration
│   └── resources/
│       ├── clients10.csv               # Sample client data
│       ├── clients100.csv              # Extended client data
│       └── reviews/                    # Product review files
└── test/
    └── java/                           # Test classes (not yet)

Setup

1. Clone the repository
2. Copy .envexample to .env and add your OpenAI API key:

OPENAI_API_KEY=your_api_key_here

3. Build the project:

mvn clean install

Dependencies

• simple-openai - OpenAI API client
• java-dotenv - Environment variable management
• jtokkit - Token counting